The newest addition to our portfolio of solutions is the Comics Identification (/v2/comics_id). This service is designed to identify comics from images. While it’s still in the early stages, we are actively refining and enhancing its capabilities.

The API detects the largest comic book in an image, and provides key information such as the title, issue number, release date, publisher, origin date, and creator’s name, making it ideal for identifying comic books, magazines, as well as manga.

This tool is perfect for organizing and cataloging large comic collections, offering accurate identification and automation of metadata extraction. Whether you’re managing a digital archive or cataloging physical collections, the Comics Identification API streamlines the process by quickly delivering essential details. We’re committed to continuously improving this service to meet the evolving needs of comic identification.

Star Wars Unlimited, Digimon, Dragon Ball, and More Can Now Be Recognized by Our System

Our trading card identification system has already been widely used to accurately recognize and provide detailed information on cards from games like Pokémon, Yu-Gi-Oh!, Magic: The Gathering, One Piece, Flesh and Blood, MetaZoo, and Lorcana.

Recently, we’ve expanded the system to include cards from Garbage Pail Kids, Star Wars Unlimited, Digimon, Dragon Ball Super, Weiss Schwarz, and Union Arena. And we’re continually adding new games based on demand. For the full and up-to-date list of recognized games, check out our API documentation.

Detect and Identify Both Trading Cards and Their Slab Labels

The new endpoint slab_grade processes your list of image records to detect and identify cards and slab labels. It utilizes advanced image recognition to return detailed results, including the location of detected items and analyzed features.

The Slab Label object provides essential information, such as the company or category (e.g., BECKETT, CGC, PSA, SGC, MANA, ACE, TAG, Other), the card’s grade, and the side of the slab. This endpoint enhances our capability to categorize and assess trading cards with greater precision. In our App, you will find it under Collectibles Recognition: Slab Reading & Identification.

Automatic Recognition of Collectibles

Ximilar built an AI system for the detection, recognition and grading of collectibles. Check it out!

Read more

New Endpoint for Card Centering Analysis With Interactive Demo

Given a single image record, the centering endpoint returns the position of a card and performs centering analysis. You can also get a visualization of grading through the _clean_url_card and _exact_url_card fields.

The _tags field indicates if the card is autographed, its side, and type. Centering information is included in the card field of the record.

Learn How to Scan and Identify Trading Card Games in Bulk With Ximilar

Our new guide How To Scan And Identify Your Trading Cards With Ximilar AI explains how to use AI to streamline card processing with card scanners. It covers everything from setting up your scanner and running a Python script to analyzing results and integrating them into your website.

Let Us Know What You Think!

And that’s a wrap on our latest updates to the platform! We hope these new features might help your shop, website, or app grow traffic and gain an edge over the competition.

If you have any questions, feedback, or ideas on how you’d like to see the services evolve, we’d love to hear from you. We’re always open to suggestions because your input shapes the future of our platform. Your voice matters!

The post New AI Solutions for Card & Comic Book Collectors appeared first on Ximilar: Visual AI for Business.

Today, I’d like to introduce a solution specifically designed for physical stores and warehouses to process their physical card collections quickly and efficiently using card scanners like those from Fujitsu. This tutorial is tailored for shop owners who need to handle large volumes of card images rapidly. We’ve developed a simple yet powerful script in Python 3 for card identification, condition assessment or grading. It also identifies comic books and reads slab labels from companies like PSA or Beckett. The script outputs a CSV file that can be easily imported into Google Sheets or Microsoft Excel. With a few modifications, it can also be adapted for use with your Shopify store or other seller tools, such as for eBay submissions. Let’s dive in and see how this tool can streamline your card-processing workflow!

Capabilities of our AI Solution for Sports Cards and TCGs

Trading Card Games

In the previous blog post, I wrote about our REST API for identifying TCGs, sports cards, and comic book covers. The TCG identification service supports more trading card games, including the most popular ones like Pokémon, Yu-Gi-Oh!, Magic: The Gathering, One Piece, and Lorcana. For some games, it can also identify the correct language version of the card or determine if it is a foil/holographic card. Additionally, for certain TCG games, the system provides links or identification numbers to the TCG Player. You can try how it works here.

Sports Cards

For sports cards, we can identify more than 5 million trading cards across six main sports categories: baseball, hockey, football, soccer, MMA, and basketball cards. Our system also supports the identification of parallel and reprint versions, with continuous improvements. Not only does it provide the best match, but it also offers alternative options to choose from.

If the trading cards are in slabs from major grading companies like PSA, Beckett, CGC, TAG, SGC, or ACE, the system can instantly identify graded cards and provide the slab company, grade, and certificate number.

All Under One API

As you can see, the functionality is complex, offering features such as bulk trading card scanning and language support, resulting in highly accurate identification. I believe that Ximilar Collectibles Recognition services are the most accurate solutions available on the market today. It is a true game-changer for card dealers, other collectors, or companies looking to be independent of third parties like CollX, Kronozio, or Card Dealer Pro, which automatically submit your cards to their marketplaces.

With Ximilar, you can handle your trading card scanning independently using our visual search technology and deep learning models. Our solutions are also designed to suit your specific needs through continuous improvements and customization. Whether you purchase, scan, analyze, search, or sell cards in bulk, our API empowers you to manage your collection without the constraints of third-party services.

How to Analyze TCG and Sports Card Scanners With AI

Step 1 – Run The Cards Through The Scanner

Enough talk! Let’s analyze the bulk of your cards. First, you’ll need a folder with images of your cards. For testing, I’ve selected a small MTG and Pokémon card subset. You can put them on your scanner via top loader (link), or individually. Most card collectors use the Fujitsu Ricoh Fi-8170 scanner, which is one of the best scanners available. It can capture both the front and back sides of the cards.

For our purposes, we will only need the front side of the cards. To avoid unnecessary costs, remove the back side images from the folder or configure your scanner to store only the front side of the cards. Some scanners, like Fujitsu, can produce scan files with names such as 19032024-0001.jpg or 19032024-FRONT-0001.jpg. You can specify the naming format for the scan files. See the following video tutorial on how to set up a Fujitsu scanner via PaperStream Capture by MaxWaxPax:

My recommendation is to use similar settings for your Fujitsu scanner as it is in the video by MaxWaxPax and create multiple profiles for sideways and top-bottom trading card scanning. Ideally set up the scanner to produce only images for the front of the cards or distinguish the images with “front” or “back” suffix in the filename. However, if you already have an unstructured collection of card images, you can fully automate the selection of images showing the front sides using our AI Recognition of Collectibles.

Step 2 – Sign Up To Ximilar Platform

Now, you’ll need an account in our App. Simply sign up with your personal or company email to get your unique API token for service authorization. Once you are in the App, copy your API key to the clipboard and save it into some file. To access the service via API, you’ll need to purchase at least a Business plan. Both tasks – getting the API key and purchasing a Business plan – can be completed in the platform’s settings in a matter of minutes.

Step 3 – Installing Python 3

Before running the script, ensure you have Python 3 installed. Some operating systems already include a version of Python, but we require at least Python 3.6. If you’re unsure, follow this tutorial on RealPython (link), which contains installation steps for Windows, macOS, and Linux:

You should be able to write in your command line, shell or terminal the similar command. Here’s mine at Mac:

michallukac@Michals-MacBook ~ % python --version && pip --version

If you don’t know how to run commands, read a short tutorial on using the terminal/shell/command line. I recommend this tutorial by DjangoGirls or watching some YouTube videos (here’s one for Windows and one for macOS). The output from the command should look similar to my example:

Python 3.9.18

pip 23.1 from /Users/michallukac/env/devel/lib/python3.9/site-packages/pip (python 3.9)

Next, you will need to install Python libraries argparse and requests via pip command:

pip install --upgrade argparse

pip install --upgrade requests

If everything passes, you’re now ready to use the script we’ve prepared to process your folder of card images!

Step 4 – Running The Script On Trading Card Games

Running the script is simple. You’ll need to use a terminal (macOS), shell (Linux), or command line (Windows), which is why we installed Python 3. Download the following file from one of these addresses:

• https://gitlab.com/ximilar-public/ximilar-client/-/blob/master/tools/process_card_scans.py?ref_type=heads

• https://gitlab.com/ximilar-public/ximilar-client/-/raw/master/tools/process_card_scans.py?ref_type=heads&inline=false

Put this file/script next to the folder (tcgscans) with your trading card images or scans and in the terminal, write the following command:

python process_card_scans.py --folder tcgscans --api_key YOURAPIKEY --collectible tcg --output results.csv --select_images all

Hitting the enter will execute the script on the folder of tcgscans, and the progress bar will be shown. The folder will analyze all the images in the folder (select_images). You can interrupt the script (it automatically stores the results every 10 images to your specified output CSV file):

Each analysis of a scan (sports card) will consume 10 credits from your credit supply in your Ximilar account. Our App lets you watch your credit consumption closely under Reports. The Business 100k Plan allows you to analyze 10,000 raw cards. If you need to analyze millions of cards per month or your entire collection at once, reach out to us, and we can offer you a bulk discount.

Step 5 – Analyzing the CSV file

Now we have our CSV file named results.csv. The CSV file contains the following fields: filename (name of the photo in the folder), status (ok or error), side (front or back), subcategory, full_name, name, year, card_number, series, set, set_code, and other additional fields.

The output format of the CSV depends on whether you analyze sports cards, TCG cards, comics, or slabs. Here is a visualization of the CSV file in Visual Studio Code:

We can import the file into Google Sheets or Microsoft Excel spreadsheet, edit it as needed, or generate printable checklists. The columns and data from the CSV can also be easily added to your Shopify product files or used for eBay submissions.

Additional information for card condition (or grading) can be added to the script via the –condition (–grading) parameter. For example, if your sports card scanner provides images with filenames such as 0001.jpg, 0002.jpg, 0003.jpg, etc., the following command will process images with odd numbering (e.g., 0001.jpg, 0003.jpg, …), identify the cards (name, card number, etc.), and also compute their condition (very good, excellent, etc.):

python process_card_scans.py --folder sportsfolder --api_key YOUR_API_KEY --collectible sport --output sport.csv --select_images odd --alternative --condition

Conclusion

With Ximilar’s AI-powered solutions, identifying and documenting your trading cards has never been easier. From trading card scanning, analyzing and organizing, to finding the current average market price, every step is streamlined to save you time and effort. I hope this guide helps you optimize your trading card workflow, making it easier to manage and showcase your collection. Happy collecting, whether it’s baseball or Pokémon cards!

The post How To Scan And Identify Your Trading Cards With Ximilar AI appeared first on Ximilar: Visual AI for Business.

Let’s take a look at what image regression is and how it works, including examples of the most common applications. More importantly, I will tell you how you can train your own regression system on a no-code computer vision platform. As more and more customers seek to extract information from pictures, this new feature is sure to provide Ximilar’s customers with the tools they need to stay ahead of the curve in today’s highly competitive AI-driven market.

What is the Difference Between Image Categorization and Regression?

Image recognition models are ideal for the recognition of images or objects in them, their categorization and tagging (labelling). Let’s say you want to recognize different types of car tyres or their patterns. In this case, categorization and tagging models would be suitable for assigning discrete features to images. However, if you want to predict any continuous value from a certain range, such as the level of tyre wear, image regression is the preferred approach.

Image regression is an advanced machine-learning technique that can predict continuous values within a specific range. Whenever you need to rate or evaluate a collection of images, an image regression system can be incredibly useful.

For instance, you can define a range of values, such as 0 to 5, where 0 is the worst and 5 is the best, and train an image regression task to predict the appropriate rating for given products. Such predictive systems are ideal for assigning values to several specific features within images. In this case, the system would provide you with highly accurate insights into the wear and tear of a particular tyre.

How to Train Image Regression With a Computer Vision Platform?

Simply log in to Ximilar App and go to Categorization & Tagging. Upload your training pictures and under Tasks, click on Create a new task and create a Regression task.

You can train regression tasks and test them via the same front end or with API. You can develop an AI prediction task for your photos with just a few clicks, without any coding or any knowledge of machine learning.

This way, you can create an automatic grading system able to analyze an image and provide a numerical output in the defined range.

Use the Same Training Data For All Your Image Classification Tasks

Both image recognition and image regression methods fall under the image classification techniques. That is why the whole process of working with regression is very similar to categorization & tagging models.

Both technologies can work with the same datasets (training images), and inputs of various image sizes and types. In both cases, you can simply upload your data set to the platform, and after creating a task, label the pictures with appropriate continuous values, and then click on the Train button.

Apart from a machine learning platform, we offer a number of AI solutions that are field-tested and ready to use. Check out our public demos to see them in action.

If you would like to build your first image classification system on a no-code machine learning platform, I recommend checking out the article How to Build Your Own Image Recognition API. We defined the basic terms in the article How to Train Custom Image Classifier in 5 Minutes. We also made a basic video tutorial:

Neural Network: The Technology Behind Predicting Range Values on Images

The most simple technique for predicting float values is linear regression. This can be further extended to polynomial regression. These two statistical techniques are working great on tabular input data. However, when it comes to predicting numbers from images, a more advanced approach is required. That’s where neural networks come in. Mathematically said, neural network “f” can be trained to predict value “y” on picture “x”, or “y = f(x)”.

Neural networks can be thought of as approximations of functions that we aim to identify through the optimization on training data. The most commonly used NNs for image-based predictions are Convolutional Neural Networks (CNNs), visual transformers (VisT), or a combination of both. These powerful tools analyze pictures pixel by pixel, and learn relevant features and patterns that are essential for solving the problem at hand.

CNNs are particularly effective in picture analysis tasks. They are able to detect features at different spatial scales and orientations. Meanwhile, VisTs have been gaining popularity due to their ability to learn visual features without being constrained by spatial invariance. When used together, these techniques can provide a comprehensive approach to image-based predictions. We can use them to extract the most relevant information from images.

What Are the Most Common Applications of Value Regression From Images?

Estimating Age From Photos

Probably the most widely known use case of image regression by the public is age prediction. You can come across them on social media platforms and mobile apps, such as Facebook, Instagram, Snapchat, or Face App. They apply deep learning algorithms to predict a user’s age based on their facial features and other details.

Needless to say, these plugins are not always correct and can sometimes produce biased results. Despite this limitation, various image regression models are gaining popularity on various social sites and in apps.

Ximilar already provides a face-detection solution. Models such as age prediction can be easily trained and deployed on our platform and integrated into your system.

Value Prediction and Rating of Real Estate Photos

Pictures play an essential part on real estate sites. When people are looking for a new home or investment, they are navigating through the feed mainly by visual features. With image regression, you are able to predict the state, quality, price, and overall rating of real estate from photos. This can help with both searching and evaluating real estate.

Custom recognition models are also great for the recognition & categorization of the features present in real estate photos. For example, you can determine whether a room is furnished, what type of room it is, and categorize the windows and floors based on their design.

Additionally, a regression can determine the quality or state of floors or walls, as well as rank the overall visual aesthetics of households. You can store all of this information in your database. Your users can then use such data to search for real estate that meets specific criteria.

Determining the Degree of Wear and Tear With AI

Visual AI is increasingly being used to estimate the condition of products in photos. While recognition systems can detect individual tears and surface defects, regression systems can estimate the overall degree of wear and tear of things.

A good example of an industry that has seen significant adoption of such technology is the insurance industry. For example, startups-like Lemonade Inc, or Root use AI when paying the insurance.

With custom image recognition and regression methods, it is now possible to automate the process of insurance claims. For instance, a visual AI system can indicate the seriousness of damage to cars after accidents or assess the wear and tear of various parts such as suspension, tires, or gearboxes. The same goes with other types of insurance, including households, appliances, or even collectible & antique items.

Our platform is commonly utilized to develop recognition and detection systems for visual quality control & defect detection. Read more in the article Visual AI Takes Quality Control to a New Level.

Automatic Grading of Antique & Collectible Items Such as Sports Cards

Apart from car insurance and damage inspection, recognition and regression are great for all types of grading and sorting systems, for instance on price comparators and marketplaces of collectible and antique items. Deep learning is ideal for the automatic visual grading of collector items such as comic books and trading cards.

By leveraging visual AI technology, companies can streamline their processes, reduce manual labor significantly, cut costs, and enhance the accuracy and reliability of their assessments, leading to greater customer satisfaction.

Automatic Recognition of Collectibles

Ximilar built an AI system for the detection, recognition and grading of collectibles. Check it out!

Read more

Food Quality Estimation With AI

Biotech, Med Tech, and Industry 4.0 also have a lot of applications for regression models. For example, they can estimate the approximate level of fruit & vegetable ripeness or freshness from a simple camera image.

For instance, this Japanese farmer is using deep learning for cucumber quality checks. Looking for quality control or estimation of size and other parameters of olives, fruits, or meat? You can easily create a system tailored to these use cases without coding on the Ximilar platform.

Build Custom Evaluation & Grading Systems With Ximilar

Ximilar provides a no-code visual AI platform accessible via App & API. You can log in and train your own visual AI without the need to know how to code or have expertise in deep learning techniques. It will take you just a few minutes to build a powerful AI model. Don’t hesitate to test it for free and let us know what you think!

Our developers and annotators are also able to build custom recognition and regression systems from scratch. We can help you with the training of the custom task and then with the deployment in production. Both custom and ready-to-use solutions can be used via API or even deployed offline.

The post Predict Values From Images With Image Regression appeared first on Ximilar: Visual AI for Business.

However, even when the results in production are good, there are areas, where we can’t simply accept black box decisions without being sure, how the AI made them. These areas are typically medicine and biotech or any other field where there is no place for errors. We need to make sure that both output and the way our model reached its decision make sense – we need explainable AI. For these reasons, we introduced a new feature to our Image Recognition service called Explain.

Training Image Recognition

Image Recognition is a Visual AI service enabling you to train custom models to recognize images or objects in them. In Ximilar App, you can use Categorization & Tagging and Object Detection, which can be combined with Flows. For example, the first task will detect all the human models in the image and the categorization & tagging tasks will categorize and tag their clothes and accessories.

Image recognition is a very powerful technology, bringing automation to many industries. It requires well-trained models, and, in the case of object detection, precise data annotation. If you are not familiar with using image recognition on our platform, please try to set up your own classifier first.

These resources should be helpful in the beginning:

• Check the Custom Image Recognition page
• Read The Basic Rules for Image Recognition Models Training
• Read Best Practices in Image Recognition Training
• Watch our YouTube tutorial on how to set up the image recognition task
• Read how to combine and chain your models with Flows

From model-centric to data-centric with explainable AI

When you want a model which performs great in a production setting and has high accuracy, you need to focus on your training data first. Consistency of labelling, cleaning datasets from unnecessary samples/labels, and adding feature-rich samples that are missing is much more important than the newest architecture of the neural network. Andrew Ng, an entrepreneur and professor at Stanford, is also promoting this approach to building machine learning models.

The Explain feature in our App tells you:

• which parts of images (features and pixels) are important for predicting specific labels
• for which images the model will probably predict the wrong results
• which samples should be added to your training dataset to improve performance

Simple Example: T-shirt or Not?

Let’s look at this simple example of how explainable AI can be useful. Let’s say we have a task containing two categories – t-shirts and shoes. For a start, we have 20 images in each category. It is definitely not enough for production, but it is enough if you want to experiment and learn.

After playing with the advanced options and short training, the result seems really promising:

Using Explain on a Training Image

But did the model actually learn what we wanted? To check, what the neural network find important when categorizing our images, we will apply two different methods with the tool Explain:

• Grad-CAM (first published in 2016) – this method is very fast, but the results are not very precise
• Blur Integrated Gradients (published in 2020) smoothed with SmoothGrad – this method provides much more details, but at the cost of computational time

In this case, both methods clearly demonstrate the problem of our model. The focus is not on the t-shirt itself, but on the head of the person wearing it. In the end, it was easier for the learning algorithm and the neural network to distinguish between the two categories using this feature instead of focusing on the t-shirt. If we look at the training data for label t-shirt, we can see that all pictures include a person with a visible face.

Explainability After Adding New Data

The solution might be adding more varied training data and introducing images without a person. Generally, it’s a good approach to start with a small dataset and over time increase it to a bigger one. Adding visually broad images helps model with overfitting on wrong features. So we added more photos to the label and trained the model again. Let’s see what the results look like with our new version of the model:

The Grad-CAM result on the left is not very convincing in this case. The image on the right shows the result of Blur Integrated Gradients. Here you can see, how the focus moved from the head to the t-shirt. It seems like the head still plays some part, but there is much less focus on it.

Both methods for explainable AI have their drawbacks, and sometimes we have to try more pictures to get a better understanding of model behaviour. We also need to mention one important point. Due to the way the algorithm works, it tends to prefer edges, which is clearly visible in the examples.

Summary

The Explainability and Interpretability of Neural Networks is a big research topic, and we are looking forward to adopting and integrating more techniques into our SaaS AI solution. AI Explainability that we showed you is only one tool amongst many towards data-centric AI.

If you have any troubles, do not hesitate to contact us. The machine learning specialists of Ximilar have vast experience with different kinds of problems, and are always happy to help you with yours.

The post Explainable AI: What is My Image Recognition Model Looking At? appeared first on Ximilar: Visual AI for Business.

Upcoming Changes in the EU Energy Labelling

The energy labels indicate the energy efficiency category the appliance falls into. In 2019, the European Union approved a new regulation setting a framework for updated energy labelling, which will come into force in 2021 and gradually replace the old system of labels. According to European lawmakers, the new system could save up to 200 billion kWh of energy, which is approximately the same amount of energy all Baltic countries spend together in a year. The first new labels are already in circulation.

Effective March 2021, sellers and manufacturers will be required to update the energy labels on fridges, washing machines, dishwashers, TVs, electronic displays, and refrigerating appliances for display purposes, followed by tyres in May, and lamps in September.

So far, the products have fallen into categories A+++ to G, which will be simplified back to A to G and the energy class of a product will be determined by higher standards. This means the appliance that was A+ in 2020 could be B or C from now on.

Re-scaling is not the only new feature, as the new labels are provided with a QR code leading consumers to the EPREL (European Product Registry for Energy Labelling) database, providing them with detailed energy and environmental information on the goods.

A Challenge for E-commerce Industry

The new regulation applies not only to retail but also to e-commerce, meaning all e-shops will be required to re-label the household appliances as well. They will be required to do so between March 1^st and 18^th.

E-shops need to identify thousands of energy labels in the product galleries and replace them with the new ones.

E-shops generally upload the energy labels as pictures into the galleries on the item pages. Due to the large amounts of images they upload every day, it is not uncommon not to have them tagged.

To ensure a smooth transition from the old label system to the new one, the physical stores will focus on the re-labelling of the displayed goods. The e-shops, on the other hand, will need to identify and replace considerable amounts of pictures in their databases at once. For instance, the largest e-shop selling household appliances in the Czech Republic Alza.cz currently offers approximately 1 200 products in the category of fridges, 500 in washing machines, 350 in dishwashers, 600 TVs, and 1 200 monitors, meaning they will need to update at least 3 850 energy labels in the first wave.

Many large e-shops also cooperate with price comparison websites, such as Heureka, that have their item galleries. For such services, the problem is a bit more complex: as a price analysis tool, the comparison website acquires its data from various sellers meaning its picture tagging or sorting is not standardised, and they have to deal with a wide range of file types and names.

Such task poses a question: what is the most efficient way to identify the old energy labels amongst other images in the product galleries in order to delete and replace them? The solution lies in the image recognition software.

Smart Solution: Image Recognition

E-shops with electronics typically upload the energy labels as images into the product galleries on their item pages and provide them to the price comparison websites. Therefore, they need software able to sort the product images, reliably recognize the old energy labels and set them aside.

Image Recognition is one of the core services of Ximilar. In principle, once you upload your images to this service, it equips them with tags and sorts them into categories. This service uses computer vision and deep learning to detect a wide range of features in the pictures. It is designed to process extensive databases of pictures in a fraction of a second.

With Ximilar App, you can develop an AI service directly for energy label recognition.

How to Use the Image Recognition on Energy Labels

If you need to identify and replace the old energy labels in your e-shop, there are two ways to use the Ximilar Energy Label Recognition service:

1. You can train your own recognition model for energy-label images. Then you can use the model as an API endpoint. Meaning, you will send images from the product gallery and get immediate feedback on whether they are or aren’t energy labels.
2. You can provide us with an export from your product image database (as image URLs or the actual files) and we will take care of the rest for you. You will get the output back in a standard CSV format.

Since image recognition is a CPU/GPU-intensive process, one of the greatest advantages of this service lies in the image database processing on our servers, whether you use the API or leave it to us. Of course, you will have a chance to test the service in the Ximilar App before you run it on your image database.

The energy label recognition with the Ximilar service is an efficient, quick, and above all, reliable way to identify the images that need to be replaced.

With Ximilar you can develop more models for energy labels recognition:

1. Reliable recognition of the old energy labels from the new ones. This might be handy in the transition period when some labels will be already replaced, but others will not.
2. Reading the actual energy class, especially from the new energy labels. The energy label change is a great opportunity to enrich your product data by this piece of information.

If you are interested, please just fill out our contact form. We are here to help!

The Image Recognition Service Makes E-commerce Easier

Whether you need to sort your catalogue into fine-grained categories, recognize pictures in product galleries, or offer similar products to your customers, Ximilar has a solution for you.

Read more in this detailed article on Image Recognition uses in e-commerce, or contact us, and we can discuss other solutions tailored to your needs.

The post Image Recognition as an Answer to New Energy Labelling appeared first on Ximilar: Visual AI for Business.

New Label Types: Categories & Tags

The post Introducing Tags, Categories & Image Management appeared first on Ximilar: Visual AI for Business.

It is visible that we are becoming experts in Fashion E-commerce and overall online commerce once the subject is image automation and data flow optimization. We either save significant expenses for our clients or improve their services for better conversion rates. Nevertheless, we are the backend guys, and we are not tangible that much to the end customer.

Redesigned Company Website

The redesign of the Ximilar website took a serious pile of man-days to create. The whole team got involved. And it is already paying off well. There is way more information about all the tools & features Ximilar offers. Spiced up with real-life use cases from many fields, including Fashion AI and E-commerce AI applications. There are examples and sample bits of code. And there is this blog to inform you about what is happening inside Ximilar.

We have also become an IBM Business Partner and expanded sales reach to Atlanta (USA), the United Kingdom and Asia. All that to be closer to you when you need a partner to help your business with the initial workings when embedding a robust Ximilar system inside your workflow.

Complex Documentation

This might seem like just a bunch of text at first look. But in reality, the documentation uncovers all the magic, all the possibilities that you get by using our universe of tools. Developers constantly update the docs, so you always have the most recent information at your fingertips. Making your life easier and supporting you in your busy day is our target.

Nevertheless, we are on email & live chat to help you anytime you require a helping hand.

New Feature: Precision/Recall table for Model

This is another critical feature to inspect the quality of your models, which should not be missed when developing a machine learning solution. We introduced a page with insight into your model in the Summer of 2018, and we are adding another advisory feature — Precision/Recall for each label. With that feature, you can verify exactly the level of reliability prediction for individual labels. The higher the value, the better your model succeeds in the prediction of these particular labels. The values have been detected from your training data, which is a random 20 % from all uploaded images in given labels.

Example: the precision of label Cats means that 85% of images that were predicted by the model to be Cats are actually cats. Low precision numbers mean that the label is too broad – many images falsely get this label, and you should probably add more training images that are NOT Cats. On the other hand, the recall of 50 % of the label Parrots means that only 5 out of 10 images that actually are parrots were recognized by the model to be Parrots. Low recall numbers mean that the training data define this label as too narrow — this label is not recognized as often as it should be, and you should add more training images that ARE parrots.

New Feature: Advanced Settings for each Task

Many of our customers tell us that models from Ximilar Recognition provide better and faster results than models of our competitors (including the big players). Knowing your data, you can now further improve the reliability of your model by selecting the right checkboxes (horizontal flip, vertical flip, rotate 90). These settings are applied randomly to your images during the training (together with other modifications that are standard in machine learning). As a result, the trained model should then be invariant to the corresponding transformation (e.g., the recognition should be independent of the vertical flip of the image).

For example, many classifiers for microscope/medical data will benefit all three to be checked, as the important knowledge on the images can be rotated in all possible ways. The common practice for basic tasks, let’s say classifying houses, is to have just horizontal flip checked (default behaviour) as you probably do not want to classify a house upside down. You can experiment with the settings as you want and see what works best for your task.

Improved & Updated Python Library

All the Ximilar Services are now behind the https://api.ximilar.com endpoint. That is why we made huge improvements to our Python library, which allows you to work with Ximilar Recognition (formerly Vize.ai), Dominant Colours, Generic Tagging & Fashion Tagging. The documentation, mentioned above, was changed to cover more knowledge, so the entire workflow of using the library is very straightforward. We still have further plans to expand this client by including more features and working with all possible endpoints.

More at https://gitlab.com/ximilar-public/ximilar-vize-api

Upcoming Features

And that is just the beginning of the year 2019. We already prepared many further features that are either requested by our customers or improve existing features to allow you to reach new horizons. These are just a few to give you a glimpse of what is coming:

• Image Tagging — or technically multi-label classification, where both the training images and the real data get more than one label/tag. A technique is often seen in stock photo agencies as photography keywords.
• Workspaces for Images and Tasks — To allow you to sort out your projects, should you have more than one.
• Improved User Interface — We are constantly iterating on the most common features.

Feel free to contact us and let us know what you are missing, or what would improve your system performance, speed or reliability. We are always on your side when it comes to reaching business targets or optimizing your expenses.

The post Major Updates for a Headstart in 2019 appeared first on Ximilar: Visual AI for Business.

Ximilar App is a powerful platform for creating your own machine-learning models. You can use the platform for free if you want to develop an AI model for classification/categorization of images, deep image tagging or detecting objects (bounding boxes) on the images. There are no coding skills needed, the system/platform is easy to use. Building your visual AI model can be done online via your browser (in the cloud) and the final model will be deployed as an API endpoint. Whether you want to recognize clothing, quality check/control on images or recognize trading cards, the Ximilar App offers you complex tools for achieving great accuracy. Ximilar is significantly cheaper than Azure AI or Google AI services (GCP, Vertex AI). The training (optimization) of the model is free! This is a huge cost saver for developing and deploying your own AI models online.

Today, I will show how to set and test a custom image classification engine using Ximilar Image Recognition. This is a step-by-step guide for training an image categorization model via the Ximilar platform. We will prepare a dataset, upload images, train the classifier and test our classifier in the web interface. We need no coding experience unless we want to build API in our project. Let’s start now.

Prepare a Machine Learning Dataset

We want the classifier to recognize cat vs dog. First, we will need 20 pictures of each cat and dog. Let’s google “cat” and “dog” and save 20 images of each. For every category, I also searched for the different breeds, so my dataset is as diverse as possible.

Upload Your Dataset to Ximilar App

We will need to create an account through app.ximilar.com. Visit the homepage, then click “Log in” and create an account through the Sign-Up form. Then you can log in to the platform. Select the Image Recognition service on the dashboard page.

We are ready to create a new task. A task is a classification engine (convolutional network model) that lets us classify our images. On the Overview page, through the Quick Actions section, we click on the “Add Task” button and pick the Categorization task. Fill the name with our classifier “Cat vs Dog”. We want to add two categories, “Cat” and “Dog”. We can always add and delete categories later. Click “Create New” to add a category on the Task page. We create two of them.

Now, we are going to “Drag and Drop” images for each category.  In Category, drag images and drop them into the Drag & Drop section. We can see 20 and 20 images uploaded in the image above. Ximilar Recognition service requires that every label has at least 20 images. We can use the “Manage Category” button on Label/Category to show images and move them from one category to another.

In this step, we can review our categories. We are ready to click on the “TRAIN” button.

A task is in training right now. It can take one to five hours, depending on the number of images and the complexity of your task. Ximilar Recognition uses transfer learning and a set of fine-tuned model architectures to reach the best possible accuracy on each task. Time to have a coffee now and wait for training to finish. After task training is finished, you will see in the model section below a list of trained models:

Our model is ready! We reached 94% accuracy on our 40 images dataset. You can view more statistics about the trained models when you click on the DETAIL button. We can now test it using Classify preview.

We trained and tested our classifier using the Ximilar web interface. This is the simplest way to build an image classification/machine learning model for photos online via a web browser. We reached 94% accuracy, which we can increase to 99.9 % by uploading more images. It is time to experiment with the huge possibilities that image classification brings. In the developers’ documentation, you can find a sample code for connection to our REST API endpoint. Here are a few more resources to help you:

• Advanced image augmentation settings for training
• Tips and best practices for training a model via Ximilar Platform
• Video tutorial for training a model

I hope you like this guide for training simple image recognition models. Contact us if you want to know more about our cloud AI platform, or sign up for free and test it by yourself. Our team can help you with your ideas and business projects.

The post How to Train a Custom Image Classifier in 5 Minutes appeared first on Ximilar: Visual AI for Business.

Earlier, I wrote a post about the difference between general and custom computer vision platforms. Today, I would like to focus on real-world use cases. Let’s dive into image recognition features planning.

Extract Patterns With AI

This might be the hardest part to recognize. I recommend training custom models for patterns because we want every image to have a pattern label. General models can detect strong patterns but do not provide patterns for every option. Having 20 pattern categories means we need only about 400 images of socks for custom model training. More information about the custom vision dataset is mentioned in this post. A custom image recognition model can be trained online via a browser, just log in to app.ximilar.com and build your own models.

Identify the Type of Sock

General vision can also detect the type of sock. I tested a few socks and got mostly “outdoor shoe” results, which are not very accurate. I prefer to spend one more hour on getting images from my e-shop database and sorting them into classes, rather than having blank spaces in my image recognition engine. Using a custom categorization model also leads to higher accuracy in classification.

Having three parameters extracted from an image and saved in the database, we are now able to create the matching logic.

Searching and Recommending Socks

Building your own image-recommending engine can be hard. Luckily, Ximilar offers a solution that can easily find products from your database of socks. We can use previous models for filtering the result on Sock type, pattern and colour.

Summary

Most computer vision tasks (models) are more complex than what one provider can deliver. It often needs some business insight, so it is necessary to take time and think about the path to our goals. Most of the solutions like Google Vision or Azure AI services are very expensive for training and deploying your custom models.

Ximilar provides easy to use and powerful solution for training and deploying your custom machine learning models for images. The training and deployment of the image recognition models are free, which saves your expenses for the development of your business idea. If you have some business ideas that require a customized computer vision solution, then contact us. We have experts and tools that can help you grow your project.

The post Add Computer Vision Features to E-commerce App appeared first on Ximilar: Visual AI for Business.