Until recently, IBM Watson Visual Recognition was primarily used to recognize an image as a whole. However, working with a picture as a whole is far from the most correct approach. Now, thanks to the new object recognition function, IBM Watson users have the opportunity to train models on images with marked objects, for their subsequent recognition on any frames.
We show how this can now be done.
If earlier using IBM Watson you could distinguish a damaged car from an undamaged one, now you can not only recognize the presence of damage, but also assess its position and size. This approach is much more informative, allowing you to make predictions about the cost of necessary repairs.
Of course, the list of applications for this functionality is much wider than a simple car integrity check. Now you can use Watson Visual Recognition to:
- Counting the number of people in lines or cars in traffic
- Identification of goods on trading shelves
- Recognition of logos in photos
- Analysis of CT scans and MRI for abnormalities
- Other tasks related to working with specific objects in pictures
You do not have to spend months on the selection and markup of data - our model is already trained on several million samples and provides a fairly high quality of prediction without any changes. If necessary, you can always retrain it so that the neural network meets the specifics of your field of activity.
Tag images and train the model on your data faster with Watson Studio
Usually training your own model to accurately recognize objects is the most difficult task when creating a computer vision system. Watson Studio speeds up this process and helps reduce time when working with large amounts of data. In conjunction with the free add-on Auto Label, you can quickly mark out all the images in the dataset.
Beginning of work
After activating and creating the Visual Recognition application in the cloud, connect it to Watson Studio and create a model in the Detect Objects window in the Custom Models section.
Upload your unallocated data to Watson Studio (you can use a JPEG, PNG or ZIP archive containing these images)
Select an image, highlight the object you want to recognize, give it a name and save. Repeat until you select all the necessary objects in this image.
Once you mark up a few images, you can train and test your model.
You can also add more images to enhance the quality of the model with the Auto Label feature that helps you tag all your data. To use this function, select all the necessary images and click on the “Auto Label” button so that Watson independently marks the data in accordance with the specified classes.
After checking the accuracy of your model, you can embed a ready-made solution in your product.
Try object recognition with IBM Watson Visual Recognition for free today!
We also want to invite you to free training seminars on IBM Watson Studio and Visual Recognition in the IBM cloud , held in November in the client center of our Moscow office.