When will Computer Vision be able to “see” exact physical objects/products?

Quick question about where you guys think the computer vision space is in terms of its capabilities and evolution. When do you think this field and the models within will be able to quickly and easily recognize exact physical products from photo and video data? For instance, instead of it detecting a “drill” it will be able to detect the exact drill it’s seeing like “Hyper Tough 8V Max Cordless Drill, 3/8 inch Chuck, Non-Removable 1.5Ah Battery with Charger, Bit Holder & LED Light.”

Thanks for taking the time and hope to hear back from you soon. Have an incredible day!

Andrew Grillo

The field of computer vision has made significant strides in recent years, and it’s already capable of recognizing and categorizing objects in images and videos. For instance, Roboflow’s object detection system can identify objects and their locations in an image, returning the coordinates of the objects it has been trained to recognize, along with a confidence level indicating the system’s certainty in its prediction source.

However, the level of specificity you’re asking about – recognizing not just a “drill” but a specific model of drill – is a more complex task. It would require a highly detailed and diverse dataset that includes images of the specific drill models you want the system to recognize. The system would then need to be trained on this dataset.

While it’s theoretically possible to achieve this level of specificity with current technology, it would require a significant investment of time and resources to gather the necessary data and train the model. It’s also worth noting that the performance of such a system would likely vary depending on the quality and diversity of the images it’s asked to analyze. For example, it might struggle with images where the drill is partially obscured or viewed from an unusual angle.

In terms of when this capability might become commonplace, it’s difficult to say. The field of computer vision is advancing rapidly, but the timeline for these advancements can be unpredictable. However, the Roboflow team is optimistic about the future of computer vision and its potential to reshape various industries source.

Hi! I’m Roboflow’s AI powered bot. This is an automated response powered using AI with knowledge from our docs, blog, knowledge base, and previous forum responses. If you find this to be a helpful answer, please mark it as the solution using the checkbox icon below. If this doesn’t solve your issue, please follow up with what you’ve tried/why, and the community will continue to chime in as usual.

This is what Toolstash is working on; might be good to look at for inspiration.

1 Like