TencentARC/UMT
UMT is a unified and flexible framework which can handle different input modality combinations, and output video moment retrieval and/or highlight detection results.
PythonNOASSERTION
Issues
- 0
- 3
Model Instability
#52 opened by SalmaMohamedElsayed - 1
about model
#51 opened by youngprogrammerBee - 1
Audio feature extraction
#40 opened by GYGWG - 1
Audio feature extraction
#48 opened by SalmaMohamedElsayed - 1
The forward method of UMT
#50 opened by czzhao-sjtu - 1
My dataset
#49 opened by anas2908 - 5
- 2
Attention map visualization
#36 opened by G-Apple1 - 1
qvhighlights/umt_base_pretrain_100e_asr.py
#46 opened by c1d2y3 - 0
The Checkpoint file requirment
#47 opened by EasonXiao-888 - 2
result of QVHighlights val set
#45 opened by EasonXiao-888 - 1
Inference code
#44 opened by hadesfgh - 2
Error. TypeError: '>=' not supported between instances of 'DataContainer' and 'int'
#43 opened by GuangtaoLyu - 2
- 6
Inference mode
#42 opened by Rj-batista - 1
Model applicability
#39 opened by gracikk-ds - 1
Query feature in TVSum highlight detection
#38 opened by GYGWG - 2
TVSum training problem
#37 opened by GYGWG - 1
音频特征提取部分的代码
#35 opened by luyanger1799 - 3
Pretraining Problem
#33 opened by Lonicer - 3
What is the horizontal coordinate of Figure 4 in the paper? What does it represent?
#32 opened by G-Apple1 - 1
- 2
results visualized
#31 opened by Yangaiei - 14
model test
#28 opened by Yangaiei - 5
- 2
- 16
feature extraction (i3d and optical flow)
#7 opened by Lvqin001 - 3
retrieve a video in real time
#26 opened by Lynneyyq - 2
automatic learning rate adjustment
#25 opened by Yangaiei - 1
- 1
save epoch problems
#27 opened by xiaohuihui-com - 1
validate
#23 opened by tangxiaochu123230 - 1
audio feature extraction
#22 opened by Yangaiei - 6
metric methods
#21 opened by oomq - 8
- 2
- 5
Hello, questions about text feature extraction。
#20 opened by Yangaiei - 7
how to align the audio feature and video feature?
#17 opened by Xuguozi - 3
How do I make my dataset
#16 opened by Yangaiei - 1
RuntimeError: CUDA error: no kernel image is available for execution on the device
#15 opened by hpppppp8 - 1
How can I annotate my own dataset?
#11 opened by Xuguozi - 1
how to visulize the results in your paper
#14 opened by wenhaoHou - 1
feature exaction
#13 opened by Xuguozi - 7
bug?? if (num_gt := sum(label)) == 0:
#10 opened by Xuguozi - 1
How to prepare the data
#9 opened by Lynneyyq - 1
.json annotation
#8 opened by Lynneyyq - 1
extract audio features
#6 opened by G-Apple1 - 1
Something seems wrong in the head.py
#4 opened by NNNNAI - 1
How to extract video features
#5 opened by Yangaiei