4 ms·
You can ask GPT-4 for bounding box of each object in json format and it’ll give it to you. Don’t know if there is understanding, but it is definitely useful
by tonmoy 3y ago
You can ask GPT-4 for bounding box of each object in json format and it’ll give it to you. Don’t know if there is understanding, but it is definitely useful
- KRAKRISMOTT 3y agoHow does it work? Does the encoder specifically pass the bounding box coordinates to the rest of the network as part of the embeddings?
- Hidios 3y agoWe don't understand how they work. We know how to make them, but much like the human brain, the very low level workings are beyond our understanding. Broadly it can be said that LLMs work by reducing uncertainty. Interestingly, human consciousness is also theorized by some to work the same way, react to the input in ways to reduce uncertainty.
- tonmoy 3y agoInternally that is probably something similar that is happening, but as far as I can tell it was never designed to specifically do that. Here is an example of the things GPT-4V is capable of: https://blog.roboflow.com/gpt-4-vision/amp/ https://blog.roboflow.com/gpt-4-vision/amp/