facebookresearch/Mask2Former
Code release for "Masked-attention Mask Transformer for Universal Image Segmentation"
PythonMIT
Issues
- 3
[BUGS] TypeError: _evaluate_predictions_on_coco() got an unexpected keyword argument 'use_fast_impl'
#212 opened by bjzhb666 - 0
Could you please let me know if anyone has successfully trained using the YouTube VIS 2021 dataset? How should the dataset be formatted?
#239 opened by Xinge-Zhou - 3
How should I fix the input size during testing?
#238 opened by klkl2164 - 0
Custom dataset registration to use a model trained on Cityscapes for semantic segmentation.
#237 opened by arslanduygu - 0
Mask loss with soft labels
#236 opened by antopost - 5
How to run this on Mac without CUDA?
#197 opened by fgolemo - 0
Poor Output image quality
#235 opened by xuyuxiu83 - 2
KeyError: "No object named 'MaskFormerHead' found in 'SEM_SEG_HEADS' registry!"
#199 opened by josyulakrishna - 0
Run in colab seems that there's a ModuleNotFoundError related to the MultiScaleDeformableAttention module.
#234 opened by henry7213 - 0
Using COCO for the dataset, what is the appropriate adjustment for learning rate if using a single GPU
#233 opened by KuKKiK9811 - 0
Prebuilt wheels provided via 3rd party repository
#232 opened by MiroPsota - 1
Unable to reproduce Cityscapes results
#202 opened by yxchng - 0
difference among different mode
#231 opened by BuRr-Lee - 1
- 2
No module named 'MultiScaleDeformableAttention', Please compile MultiScaleDeformableAttention CUDA op
#222 opened by kanishkanarch - 0
loading swintransformer
#228 opened by BuRr-Lee - 0
why swin accept different input size
#227 opened by sleepat-11 - 1
how use custom pre-trained backbone in mask2former
#226 opened by sleepat-11 - 0
batch_size doesn't affect evaluation
#225 opened by SuroshAhmadZobair - 1
- 1
- 0
As for training, how long does it take?
#223 opened by zckcola - 1
Same pixel for 2 instance mask
#218 opened by Mythili-kannan - 0
- 0
How to understand the output of different tasks
#220 opened by Zhong-Zi-Zeng - 2
A question about pixel decoder
#205 opened by LiJiahang617 - 0
- 0
Correct way to deal with empty masks
#215 opened by kaitolucifer - 0
how to use masked attention
#214 opened by zhangxiaoxuan39 - 0
Typo in MODEL_ZOO.md
#213 opened by QihangYang - 2
The last dimension of predicted logits discarded in inference but used in training
#200 opened by questionstorer - 0
Windows Support
#211 opened by vedanshthakkar - 0
- 0
Problem with cell segmentation
#209 opened by jiaqixuac - 0
Using FPN as Pixel Decoder
#208 opened by levan92 - 3
Is Mask2Former really using AMP?
#188 opened by yxchng - 1
- 1
"RuntimeError: CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1."
#207 opened by rockywind - 1
- 0
Mask2Former Code vs Weights Licenses
#206 opened by alzaia - 0
Mask2Former Model Weights License
#204 opened by SprAJR - 0
How to get multiscale inference results?
#201 opened by yxchng - 1
- 1
- 0
- 0
Very precise borderline.
#195 opened by sushilkhadkaanon - 0
Demo prediction gets killed in Jetson Nano
#194 opened by kanishkanarch - 0
Force number of masks for the panoptic segmentation
#193 opened by guist - 1
Where can I find the class definitions? The Cityscapes dataset has 30 classes now.
#192 opened by Chan0081 - 0