hkchengrex/Mask-Propagation
[CVPR 2021] MiVOS - Mask Propagation module. Reproduced STM (and better) with training code :star2:. Semi-supervised video object segmentation evaluation.
PythonMIT
Issues
- 1
YouTubeVOS dataset cannot be downloaded
#47 opened by theRain0450 - 0
About top-k filtering
#46 opened by csustYyh - 1
license file missing
#43 opened by sparshgarg23 - 4
About Fig.6 again
#45 opened by csustYyh - 7
Confusion about Fig.6
#44 opened by csustYyh - 1
Does the full model means s012 model?
#42 opened by zhouweii234 - 5
- 2
about train stage
#41 opened by lsy-dot - 4
metrics results of test dataset
#40 opened by xwhkkk - 1
- 2
thin-plate-spline question
#37 opened by cheind - 1
Which YouTubeVOS do you use to train?
#36 opened by zhouweii234 - 1
- 1
Does the kernelized memory need training?
#35 opened by zhouweii234 - 14
Question about result on DAVIS16
#33 opened by yongliu20 - 31
About BL30K
#29 opened by longmalongma - 4
J&F performance on BL30K
#32 opened by vateye - 8
- 6
How do I get the penultimate frame?
#30 opened by longmalongma - 2
some question about codes
#27 opened by longmalongma - 1
About BK30K
#28 opened by longmalongma - 2
utf-8''BL30K_c.zip里面包含了 utf-8''BL30K_c.tar文件
#26 opened by longmalongma - 4
如何取所有内存帧中的前一帧?
#25 opened by longmalongma - 1
Does --benchmark speed up training?
#24 opened by longmalongma - 2
Is it possible to further improve the accuracy by pre-training with other static images and virtual video?
#23 opened by longmalongma - 2
about batch size
#22 opened by longmalongma - 10
- 4
- 2
Backbone uses Resnet50?
#20 opened by longmalongma - 1
- 4
- 3
Do I need to manually merge UTF-8 "BL30K_A, B, and C into the BL30K directory after the data set BL30K is downloaded?
#16 opened by longmalongma - 8
- 2
What if I didn't use BL30K?
#14 opened by longmalongma - 2
How do I get quantitative results after testing on the DAVIS 2016 dataset?
#13 opened by longmalongma - 4
RuntimeError: CUDA error: out of memory
#11 opened by longmalongma - 1
How to skip the pre-training of static images and BK30L and go directly to VOS training?
#12 opened by longmalongma - 1
only mask propagation?
#10 opened by longmalongma - 4
- 3
Has anyone experienced CUDA out of memory?
#8 opened by Meryl-Fang - 19
subprocess.CalledProcessError
#3 opened by longmalongma - 4
How to install thinplate manually?
#7 opened by longmalongma - 3
about BL30K
#6 opened by longmalongma - 20
The server remained unresponsive for a long time when I try to train your model.
#4 opened by longmalongma - 9
- 4
How to install thinplate?
#2 opened by longmalongma