issues
search
OpenGVLab
/
VideoMAEv2
[CVPR 2023] VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking
https://arxiv.org/abs/2303.16727
MIT License
444
stars
45
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
on the tad features extraction, is image normalization required?
#62
auzxb
opened
2 weeks ago
0
Knowledge distillation code
#61
rixejzvdl649
opened
2 weeks ago
1
Zero-shot evaluations on downstream datasets
#60
XuecWu
closed
1 day ago
0
CLS token
#59
Alsalivan
opened
1 month ago
0
Finetuning with more than 16 frames
#58
CSLR-research
opened
1 month ago
0
Code implementation of model inference
#57
XuecWu
opened
1 month ago
0
Impact of Something Something and Kinetics during Unlabeled Pre-training
#56
edessa
opened
2 months ago
0
fine-tuning AVA dataset for spatiotemporal detection
#55
Young-eng
opened
3 months ago
0
2333
#54
justzhanghong
closed
4 months ago
0
The parameter grad_norm appears to be inf and then nan when input resolution is 112*112 during the pre-training on VIT-Small backbone
#53
DragonWang-cell
opened
5 months ago
1
What should I do if I want to get the features of ActivityNet-1.3?
#52
hongminglin08
closed
3 months ago
7
Where to find the script of finetuning on 'Temporal action detection' task?
#51
Leo-Yuyang
closed
5 months ago
6
Could you provide ActivityNet 1.2 and ActivityNet 1.3 features extracted by videomaev2 ?
#50
Value-Jack
closed
3 months ago
1
Could you provide features for ActivityNet 1.2 and ActivityNet 1.3 features extracted by videomaev2 ?
#49
Value-Jack
closed
3 months ago
0
Finetuned smaller models
#48
vpfuentealba
closed
3 months ago
1
Apply VideoMAEV2 to other directions.
#47
dufan175
closed
3 months ago
2
Initialize student model's weights
#46
duytue-kite
closed
3 months ago
3
Pretrained smaller models availability
#45
ganzobtn
closed
3 months ago
1
VideoMAEv2-L Weights/Checkpoints
#44
CarrotPeeler
closed
3 months ago
1
Extracting Features from Frame Level Data
#43
n1791348200
closed
3 months ago
1
adjust of the dataset code.
#42
JerryFlymi
closed
3 months ago
0
adjust the hybrid dataset code for pretrain.
#41
JerryFlymi
closed
8 months ago
0
Turning VideoMAEv2 into a next-frame prediction model
#40
IoSonoMarco
opened
8 months ago
1
Error when running runclass_finetuning.py
#39
Ravindu-Yasas-Nagasinghe
closed
3 months ago
3
Failed to achieve claimed accuracy using vit-b on K400 dataset
#38
HaoRanLyu
closed
9 months ago
6
Pre-train Action recognition videoMAE model on UCF101
#37
Mr-MeerMoazzam
closed
3 months ago
12
Visualization Script
#36
kfirgoldberg
closed
3 months ago
2
Apply for the model weight of 'vit_g_hybrid_pt_1200e_k710_it_k400_ft'
#35
jinyucn
closed
3 months ago
1
Request for the training script for VideoMAE-V2-Base
#34
qinghuannn
closed
3 months ago
3
Starting the pretraining from checkpoint..
#33
SushantGautam
closed
10 months ago
2
Clarification on published logs.
#32
SushantGautam
closed
10 months ago
1
No module named 'petrel_client'
#31
SeeeeShiwei
closed
11 months ago
1
How to train my own dataset?
#30
wang9danzuishuai
closed
11 months ago
2
Pretrained Action Detection on AVA-Kinetics model weights
#29
girmaji08
closed
3 months ago
19
预训练的大模型下载地址希望给一个国内能访问的
#28
bimver
closed
11 months ago
2
In datasets/build.py > def build_dataset > should the "anno_path" for mode = 'test' be (args.data_path, 'test.csv') instead of what it is currently (args.data_path, 'val.csv')?
#27
yerx
closed
11 months ago
1
Could you provide the "misc/label_710to710.json" file?
#26
yerx
closed
11 months ago
1
could you please provide the weights of VideoMAEv2 pre-trained on Kinetics-400?
#25
fmthoker
closed
10 months ago
2
Unable to load the distilled model weights provided in the model zoo
#24
druefena
closed
1 year ago
0
I find that there seems to be some strange things in the evaluation of model.
#23
leexinhao
closed
1 year ago
7
Do you have the finetuned checkpoints for UCF101?
#22
yerx
closed
1 year ago
2
The hyperparameter Settings in the script seem to be inconsistent with those in the paper
#21
leexinhao
closed
1 year ago
4
(Feature request) Batched feature extraction
#20
christian-matroid
closed
3 months ago
18
[Doc] Release TAD Features
#19
congee524
closed
1 year ago
1
Wonder more pretrain scripts and results
#18
LinB203
closed
1 year ago
2
[Doc] update bibtex of cvpr
#17
congee524
closed
1 year ago
0
[Feature] Support pretraining with PyTorch 2.0
#16
congee524
closed
1 year ago
0
你好!再向你请教一个问题,就是我把部分模块冻结不更新参数的时候,跑的V2版本的vit_b_k400_ft.sh,batch size设置为4的时候一个epoch训练时间为1小时20分钟,batch size设置为8的时候一个epoch训练时间也为1小时左右,batch size设置为32的时候一个epoch训练时间也为1小时左右,请问这是正常现象么,就是batch size增大4倍的时候,每一个step时间也会增大四倍,然后一个epoch的总时间就不怎么变化,但无论batch size是4,8,还是32,GPU利用率好像都是满的(GPU-Util Compute M.这一列),请问我这里成倍数增加batch size而不能成倍数减少训练时间是正常的吗,目前batch size为4和8都能完整训练十个epoch,但是为32的时候会报错RuntimeError: DataLoader worker (pid 34621) is killed by signal: Killed.
#15
DragonWang-cell
closed
1 year ago
7
你好!我跑的V2版本的vit_b_k400_ft.sh,最终测试final_test需要20个小时,如下面所示,然后我又跑VideoMAE的final_test,发现也差不多那么久,但是我记得之前跑测试就俩小时左右啊,这是怎么回事啊,是我记错了么,修改了一下午v2版本的然后还是这样,突然找不到原因了
#14
DragonWang-cell
closed
1 year ago
2
你好!请问可以提供ViT-base蒸馏模型finetune的script或者提供ViT-base的普通模型吗?非常感谢!!!我的邮箱是2256380854@qq.com
#13
DragonWang-cell
closed
1 year ago
9
Next