-
I have run the fusion.m.And found that this code may need very long tme.I ran about 10 hours and it still in epoch 01 : 600/994. The number of epoch seems to be 300.So this program need about 150 days…
-
参考这个文档进行打架识别的复现：
https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.5/docs/advanced_tutorials/customization/action_recognotion/videobased_rec.md
在执行以下抽帧命令是出现报错：
python data/ucf101/ext…
-
I just trained for 45 epochs and set LR 0.01 for UCF101, but I got the really terrible result. Firstly the training loss is stable around 4, but cannot get lower(because the dataset is small, it shoul…
-
Hi
When I set batch size as 40 for hmdb51 or 80 for ucf101 in terms of iframe, the training would stop showing "cuda: out of memory". I have to reduce batch size to 20 or 30, but the training proces…
-
## Summary
Include model action/activity/interaction recognition (video).
## Datasets
- [Kinetics400](https://paperswithcode.com/dataset/kinetics-400-1)
The dataset contains 400 human action classes,…
-
After I converted the hmdb51 and ucf101 data sets according to the "reencode.sh", some mp4 videos could not be recognized during training, but some mp4 videos success.
And the unrecognizable videos w…
suki7 updated
2 years ago
-
ucf101/train_256.sh. 8V100, batchsize = 6*8=48, the loss is decreased to 0.09, but the sample result is bad
![sample_000](https://github.com/PKU-YuanGroup/Open-Sora-Plan/assets/7641281/7bc6c53c-d4d3…
-
Hello,
I am using compute_volume_mean.sh and my exact command is
GLOG_logtostderr=1 ../../build/tools/compute_volume_mean_from_list.bin list_HAR_train.out 16 128 128 25 HAR_train_mean.binarypr…
-
I'm trying to fine tune ResNet-34 for UCF101 dataset with this command
python main.py --root_path ~/Downloads/data --video_path UCF101/UCF101_jpg --annotation_path UCF101/ucfTrainTestlist/ucf101_01…
-
Thanks for your great work, but when I tried test reuslt, I met the following problem, Do you have any idea?
Thanks a lot @vra
python action_tools/eval_linked_results.py --imdb UCF101_RGB_1_FLOW_5…