jeffreyyihuang/two-stream-action-recognition

Issues

• Cannot unzip ucf101_jpegs_256.zip

  #83 opened 5 months ago by Nicky66667
  0

• pickle文件怎么生成的，具体都要包括什么内容啊，谢谢

  #80 opened 3 years ago by syfffffff
  1

• 博主你好，GitHub上公布预训练模型无法下载，能不能发我邮箱一份1092940968@qq.com

  #54 opened 6 years ago by zhangjia3
  18

• spatial_cnn overfitting

  #57 opened 6 years ago by Robinxin123
  5

• TypeError: new(): invalid data type 'str'

  #82 opened 2 years ago by wtt55
  0

• UnicodeDecodeError: 'ascii' codec can't decode byte 0xd0 in position 0: ordinal not in range(128)

  #47 opened 6 years ago by fcxmcc
  2

• 请问frame_count.pickle文件是如何生成的，谢谢

  #60 opened 5 years ago by JWZhao-uestc
  2

• How to generate pickle files? What should be included? Thank you

  #81 opened 3 years ago by syfffffff
  0

• Does this code have no data preprocessing (generating RGB image and optical flow diagram)

  #79 opened 3 years ago by syfffffff
  0

• 请问，此代码是不是先运行spatial_cnn.py在运行motion_cnn.py,最后在运行average_fusion.py,感谢

  #78 opened 3 years ago by syfffffff
  0

• A problem I meet in spatial_cnn and spatial_dataloader

  #77 opened 3 years ago by inkplatform
  0

• how can i use this code in Win10?

  #56 opened 6 years ago by goforfar
  1

• 写的太混乱了

  #76 opened 4 years ago by tengfei988
  0

• pretrained model

  #75 opened 4 years ago by wuyifujia
  0

• How can I use flownet2.0 method to generate 2-channel optical flow image ?

  #74 opened 4 years ago by anzheng13
  0

• About optic-flow？

  #73 opened 4 years ago by QySongxx
  0

• How did you get the performance you reported?

  #72 opened 4 years ago by khm159
  0

• Improvement on motion-cnn result: 84.1% on split-1, with VGG-16

  #39 opened 6 years ago by gaosh
  42

• Excuse me, what is the operating environment of this experiment?Thanks

  #64 opened 5 years ago by hsd201809
  1

• The environment?

  #68 opened 5 years ago by Quxyz
  1

• Excuse me, is there a tensorflow version?

  #71 opened 5 years ago by lalala666-creator
  0

• Please tell me how the frame_count.pickle file is generated. Thank you. I want to use hmdb51

  #70 opened 5 years ago by honest2017
  0

• How can I generate 2-channel optical flow image and save its x, y channel as .jpg image?

  #42 opened 6 years ago by fairyzhxz
  1

• choose Only Testing , but still train data

  #41 opened 6 years ago by sxzy
  31

• what is the purpose of val_sample19 ?

  #69 opened 5 years ago by ratnesh1729
  0

• Missing License

  #65 opened 5 years ago by PatrickBue
  1

• If I take 64 x-channel and 64 y-channel optical flow images as input to motion cnn, should I make the out channel num of conv1_custom bigger?

  #66 opened 5 years ago by ljmiao
  0

• 关于视频数据的预处理

  #50 opened 6 years ago by DoubleYing
  2

• Error while testing.

  #63 opened 5 years ago by astha0899
  0

• H

  #61 opened 5 years ago by JWZhao-uestc
  0

• HandStandPushups 在代码中多次单独处理，为什么？谢谢

  #62 opened 5 years ago by JWZhao-uestc
  0

• The dataset mismatches with the code

  #59 opened 6 years ago by liushaocan
  1

• The running time

  #58 opened 6 years ago by Songyyan
  0

• Missing "v" folder in "UCF101/tvl1_flow/"

  #36 opened 6 years ago by sunnydps32
  3

• What is the accuracy reported on? Validation?

  #53 opened 6 years ago by duygusar
  0

• Specifications for virtual environment

  #51 opened 6 years ago by duygusar
  2

• no CUDA-capable device is detected (only for motion network)

  #52 opened 6 years ago by duygusar
  1

• There are some repeat video id in testlist and trainlist

  #49 opened 6 years ago by Airror
  0

• choice of making flow tensor

  #48 opened 6 years ago by ratnesh1729
  1

• Is frame-wise min-max normalization used for flow images?

  #46 opened 6 years ago by LaiSongxuan
  0

• How many Gpus you use？

  #45 opened 6 years ago by zeal-up
  0

• Can you share specification of pc?

  #44 opened 6 years ago by Zolza09
  0

• There are problems in the pre training model！！！

  #43 opened 6 years ago by study-renjie
  1

• about the train and test data

  #40 opened 6 years ago by sxzy
  0

• Pre-trained models are corrupted

  #38 opened 6 years ago by rkdasari
  1

• the problem of the pre-trained model

  #37 opened 6 years ago by Tylerjoe
  3

• the dataset of flow are damaged files

  #32 opened 6 years ago by Tylerjoe
  1

• UCF_101 is not defined

  #33 opened 6 years ago by sunnydps32
  1

• UCF101_splitter not defined

  #34 opened 6 years ago by sunnydps32
  1

• some problem of pretrained models？

  #35 opened 6 years ago by sdd9465
  0