GZ-YourZY

Is it possible to publicize the code related to sparse matching? If yes, is it appropriate to make it public?

1

Recently, I was doing a work related to sparse matching, and calculated the Euclidean distance between each patch and other image patches using dinov2 weights, and found that the result...

Each training session ends with only one epoch of training

5

I've built my dataset up and tried to fine tune the dinov2, but it only executes one epoch per training and then ends, I'd like to know which parameter this...

Pre-training weights and model loading show mismatch

1

I used the pre-training model you posted and the pre-training model loaded by default and it shows a model mismatch, why is that? Traceback (most recent call last): File "train.py",...

I hope to use SAM's own prompter to input point and label for inference

1

## Description I wish to extend the multimodal library and the code in the Convlora example To fine-tune SAM through ConvLoRA, I hope to use the prompter of SAM itself...

This work is interesting, but I would like to know whether it is possible to manually perform prompts after using LoRA to fine-tune SAMed to avoid differences in results caused...

Does your method require training?

1

I tried your method on my data and found that the matching was unsatisfactory, so I was wondering if your method requires training. Because I don't see the training code.