3 ms·
A nice feature of CNNs is that you can change the resolution at inference time without retraining. For instance, when the user plugs in a camera with a differen
by jacobgorm 1y ago
A nice feature of CNNs is that you can change the resolution at inference time without retraining. For instance, when the user plugs in a camera with a different aspect or decides to the change the orientation of his phone from landscape to portrait. It is not clear to me if VITs can support aspect or resolution changes without any retraining?
- lava_pidgeon 1y agoCan you elaborate? In my experience it is the opposite: CNNs are highly depend on the input tensor shapes thus resolution change need even an architectional change. While resolution changes in ViT lead to more tokens, a ViT model can handle that (for image classification e.g. you always take the CLS token, Segmentation maps and similar task have the same output as in the input).