google-deepmind kinetics-i3d issues

google-deepmind / kinetics-i3d

Convolutional neural network model for video classification trained on the Kinetics dataset.

Apache License 2.0

1.75k stars 462 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

How should I normalize the framerate?

#77 skaws2003 closed 5 years ago
2
Can you tell me the hyperparameters in hmdb51?

#76 seominseok0429 opened 5 years ago
2
I3D crashes at the last average pooling 3D if the network has a shape < 2 in its temporal dimension

#75 mourad1081 closed 5 years ago
2
ResourceExhaustedError when running the file evaluate_sample.py with my own video

#74 mourad1081 closed 5 years ago
3
Can you provide the score files as well as the training and testing lists on hmdb51 and ucf101 ?

#73 ZealACMer opened 5 years ago
0
about frames

#72 ghost opened 5 years ago
1
How to select clip temporally?

#71 hollowgalaxy closed 5 years ago
1
out of memory

#70 geminikq opened 5 years ago
1
what are input_arrays and output_arrays for the model

#69 samaa opened 5 years ago
0
Difference in precision of the probabilities & logits obtained

#68 sparsh-b opened 5 years ago
0
Problem with dependencies

#67 v-kogr closed 5 years ago
7
extract features

#66 123liluky opened 5 years ago
1
What minimum amount of GPU is needed to finetune pretrained ? 8gb GPU is enough? or i need 12 gb GPU?

#65 siddharth2022 opened 5 years ago
3
What is the difference of I3d and 3d ConvNet?

#64 cemilzalluhoglu opened 5 years ago
9
Localization of Activity

#63 pcarun27 opened 5 years ago
2
How to use the kinectics pre-trained I3D as Feature Extractor for my own data

#62 rishabh2301 closed 5 years ago
1
how to test model on my sample video

#61 alighofrani95 opened 5 years ago
4
Fusion of RGB and flow models?

#60 ir0nt0ad closed 5 years ago
1
pre-process videos to 25 FPS

#59 bomri opened 5 years ago
6
Correct links to Kinetics dataset

#58 anuragarnab opened 5 years ago
3
how to use the pretrained model

#57 ilovekj opened 5 years ago
0
Kinetics-600 checkpoints for optical flow

#56 gaurvigoyal closed 5 years ago
4
META DATA FILES OF UCF-101 DATASET FOR ACTION RECOGNITION

#55 abhisheksurya578 opened 5 years ago
0
where can I get the .pb file for the model

#54 ajcwhiteintel opened 5 years ago
0
The difference between the 3D Inception Module and the 2D.

#53 zeal-up opened 5 years ago
0
code to load 2D parameters (pretrained on imagenet) to 3D model

#52 emergencyd opened 5 years ago
3
Missing data in Kinetics 400

#51 jhagege opened 5 years ago
1
Dependencies

#48 zwenaing opened 5 years ago
1
5k steps on UCF-101 => steps in terms of?

#47 MStumpp opened 5 years ago
4
train from scratch on ucf101 dataset

#46 leviswind opened 5 years ago
10
train on my own dataset

#45 95xueqian opened 5 years ago
14
How to test using the whole videos?

#44 zeal-up closed 5 years ago
4
What is the minimum hardware requirement to fine-tune I3D on new dataset?

#43 stephencheung16 closed 5 years ago
3
depth data

#42 sanaz-rezaei opened 6 years ago
1
LiteFlowNet over OpenCV TV-L1 optical flow algorithm. Can we ?

#41 estathop opened 6 years ago
5
Feature Extraction from Last Global Average Pooling Layer

#40 estathop closed 6 years ago
7
Regarding the 2 dimensions of the Optical Flow

#39 cdjkim opened 6 years ago
2
Has anybody reproduced the results on UCF101 and HMDB

#38 XudongLinthu opened 6 years ago
2
Instructions how to train the model?

#37 jillelajitta opened 6 years ago
0
Regarding TV-L1 Optical Flow

#36 november07 closed 6 years ago
4
is there have pytorch version?

#35 cuixing158 opened 6 years ago
1
Data-preprocessing for kinetics-400

#34 YongyiTang92 opened 6 years ago
17
Parameters for finetuning on UCF101 and JHMDB

#33 jiaozizhao opened 6 years ago
1
How can l adapt evaluate_sample.py for hundred of samples ?

#32 Benjiou opened 6 years ago
1
no

#31 pinkfloyd06 opened 6 years ago
0
*.npy files include RGB and flow ，which include 79 frames respecitvely.but each rgb frame to a flow frame?

#30 daixiangzi opened 6 years ago
0
automatic sign language recognition.

#29 AtaaEddin opened 6 years ago
2
Is it better to train from scratch on Kinetics-600?

#28 WuJunhui opened 6 years ago
36
What regularizers do you use when training?

#27 WuJunhui opened 6 years ago
1
How to generate .npy files for different videos?

#26 AdarshMJ opened 6 years ago
3

Previous Next