jx-zhong-for-academic-purpose / GCN-Anomaly-Detection

Source codes of our paper in CVPR 2019: Graph Convolutional Label Noise Cleaner: Train a Plug-and-play Action Classifier for Anomaly Detection
234 stars 44 forks source link

About select high confidence snippets #6

Closed poweryin closed 5 years ago

poweryin commented 5 years ago

Hi,Thanks for your great work.I downloaded the extract feature code. I have some doubt. Firstly, the c3d_iter_1000.caffemodel file format you provided is a pcx image, not a binary file. Secondly, in the cross-entropy error of direct supervision, In the paper ,H represents a set of high-confidence snippets. eg.C3D net ,So does the max value of |H| is the value 1600*60%?why does H plus absolute value mean ? I don't understand words in the paper"Due to the limited memory of GPUs,we at most sample 1600 high-confidence snippets with not more than 8 neighbors respectively in a video".Is an epoch selecting a 1600 high confidence snippets?what‘s the 8 neighbors mean? Looking forward to your reply Best wishes

jx-zhong-for-academic-purpose commented 5 years ago
  1. c3d_iter_1000.caffemodel is a binary file, and it can be shown by Netron: image
  2. H plus absolute value means the cardinality of the set H. For more details, you can visit https://en.wikipedia.org/wiki/Cardinality.
  3. Your remaining problems will be addressed after our training codes is released.
poweryin commented 5 years ago

Thanks for your reply,I get it