JJBOY/BMN-Boundary-Matching-Network
A pytorch-version implementation codes of paper: "BMN: Boundary-Matching Network for Temporal Action Proposal Generation", which is accepted in ICCV 2019.
PythonMIT
Issues
- 7
BMN model 没有完成分类任务吗?只完成proposal generation ?
#48 opened by fangxu622 - 7
Anyone reproduce the results on THUMOS14?
#31 opened by Jockey721 - 1
能不能请你分享THUMOS14的实验代码呢?
#43 opened by pengjinqiang - 0
What is the input of the BMN model?
#54 opened by omrastogi - 5
the pth file cannot be decompressed
#14 opened by Light-- - 3
- 0
How to do CLASSIFICATION TASK with BMN work?
#53 opened by xianguo-dev - 0
Training speed
#52 opened by VitalyyBezuglyj - 3
TEM loss is not really decreasing
#44 opened by yt2639 - 1
some questions...
#51 opened by jojolee123 - 0
这里的AR@1 AR@5啥意思呐,似乎对不上号
#50 opened by Jerrynogjf - 10
any classification related to the module?
#20 opened by sijun-zhou - 0
The number of temporal proposals.
#47 opened by ByZ0e - 0
- 0
featureFrame
#42 opened by M-Mokari - 0
why conv1d group is 4?
#41 opened by hhhhnwl - 13
Expected Performence
#6 opened by frostinassiky - 3
temporal dimension ???
#40 opened by Aries5522 - 1
- 0
The Proposal Evaluation Module is not Learning
#38 opened by sauradip - 1
- 2
About the label and the best_loss
#37 opened by doudou123456 - 0
How to calculate AUC
#36 opened by FuNian788 - 4
csv_mean_100/v_nt4Ag91306U.csv does not exist
#35 opened by lucky-23 - 8
csv_mean_100/v_t_D9MYkEPEo.csv' does not exist
#11 opened by Light-- - 8
pem loss nan?
#23 opened by leemengxing - 4
Fix BMN model's structure
#33 opened by FuNian788 - 0
channel_out in PEM x_1d_p
#34 opened by lyx190 - 10
about the length of temporal dimension?
#24 opened by semchan - 4
Non-ASCII character '\xe9' in file main.py
#13 opened by Light-- - 1
can train model without gpu?
#26 opened by yuningkang - 2
How does the author turn unequal video into equal length? rescaled the feature length of all videos to same length 100,
#28 opened by Niclaus233 - 1
- 2
Question about how to get the weight
#29 opened by WhoSJackie - 2
- 0
can train model
#25 opened by yuningkang - 2
about feature extraction
#10 opened by sunnymoon155 - 5
nan tem_loss in the middle of the traing
#22 opened by yangsusanyang - 0
Please advise your cuda version
#21 opened by yangsusanyang - 1
About video_info_new.csv file
#19 opened by BlueDing101 - 2
The training takes 10 hours to finish?
#16 opened by Rheelt - 0
Visualize Actionness probability Sequence
#17 opened by pranavgundewar - 1
Code for activity detection
#9 opened by SilvioGiancola - 1
Training parameters
#8 opened by BlueDing101 - 1
- 4
why the training of BMN get stuck?
#5 opened by Zhuysheng - 1
Getting lower results
#4 opened by ophirazulai - 2
- 4
PEM loss=NaN
#2 opened by yklilfft - 1