SegVit
SegVit copied to clipboard
zbwxp
Official Pytorch Implementation of SegViT: Semantic Segmentation with Plain Vision Transformers
Hi, I wondered if I can use your codebase to reproduce your best result (IoU of 58.0 on ADE20k) from the SegViTv2 paper, as it seems that the code only...
Thank you for uploading the code for your interesting project. We are trying to retrain your SegVit code using our dataset. Is it possible to change the network so that...
Hello, https://github.com/zbwxp/SegVit/blob/85616388cd8983e2d4daafd971540b210e956666/configs/_base_/models/seg_vit-b16.py#L24C9-L24C23 I found that you set with_cls_token=False. Is this means that ViT encoder in your model do not use class token, and use it for the decoder(ATM)?
https://github.com/zbwxp/SegVit/blob/85616388cd8983e2d4daafd971540b210e956666/decode_heads/atm_single_head.py#L277 semseg = mask_cls.softmax(-1) @ mask_pred.sigmoid() which will result in: **1>= semseg >= 0** When we compute [loss_func](https://github.com/zbwxp/SegVit/blob/85616388cd8983e2d4daafd971540b210e956666/losses/criterion.py#L58), we set semseg as a **logits**, and compute prob by semseg.sigmoid() If...
Nice work! I am mostly interested in your recent work _SegViTv2_, which can be easily extended to semantic segmentation under the continual learning setting. Where is the corresponding code? Look...
Hi, I wonder if there is specific torch/cuda version requirements for running segvit. I have following packages installed and I get error for mmcv: mmcv-full==1.4.4 mmsegmentation==0.24.0 torch==2.4.1 cuda version=12.4 mmcv...
Hi! I'm new to working with Vision Transformers and currently exploring various approaches to image segmentation. While reading the paper, I found the approach taken here quite interesting. However, I'm...
