HITerStudy

Results 7 comments of HITerStudy

When can we expect the training code for OWL-ViT?

@mjlm hello, thanks for your fancy work! When the all of codes can be updated? Please offer the full codes which include the configs for experiments published in the paper....

The paper is inconsistent with the size of the code parameters

> Hello there! Very happy to use your code, I have a question looking forward to your reply! > The same structure of DenseNet, the parameters of the pytorch version...

Will you provide the training config file for ViT-L (66.0 AP)?

The number of encoder layer number and decoder layer in Co-DINO for 66.0AP is 6? these is no any description about this part in the paper.

Will you provide the training config file for ViT-L (66.0 AP)?

> @HITerStudy We find the performance saturates when using more than 6 encoder or decoder layers for larger models (e.g., Swin-L). So we use 6 layers by default. Thanks for...

Will you provide the training config file for ViT-L (66.0 AP)?

how to implement the TTA used for the ViT-L(66.0AP), could you describe some details? Thank you.

A question about CLIP_as_supervision

Thanks for your wonderful job! I also meet the same problem, please offer the missing files in CLIP_as_supervision as possible as soon! Thanks!

When to release the code?

@attn4det how to combine the group_detr with DINO, how to process the dn and mix-selection part? Can you tell some details, thanks!