how to test my own custom video？

[HuitMahoon]

i have trained and tested with your datasets h5 file .And i just want to test my own video file ,like 'my_video.mp4'. how can i transform it into h5 file ,and just use your code "python main.py -d datasets/my_own_video.h5 -s datasets/summe_splits.json -m summe --gpu 0 --save-dir log/summe-split0 --split-id 0 --evaluate --resume path_to_your_model.pth.tar --verbose --save-results"

[SinDongHwan]

@HuitMahoon Hi, you should create 'features', 'fps','picks','change_points','n_frame_per_seg' to test your dataset.

• features : feature of every 15th frames.
• fps : frame per second.
• picks : 15
• change_points : shot(seg) change point( you can get change points using KTS )
• n_frame_per_seg : number of frames per segment.

refer my repository.

Good Luck!!

[TriLoo]

I have a question, the features extracted from resnet is not same as the downloaded .h5 data, so, would the data distribution mismatch between two feature extractor network cause not good results? thanks @SinDongHwan

[MHGCC]

Have you solved this problem now？Or is there a more detailed video preprocessing process to generate HDF5 files？

[SinDongHwan]

Hi, @TriLoo, @MHGCC. I had gotten different key points, too. so I've not solved yet. I think that you will good key points if use a high accuracy model for extracting features.

dear @MHGCC. if you want to generate a HDF5 file of yours dataset for testing, you can refer my repository. but you want to make it for training, you will create "scores" or "summary".

[anaghazachariah]

You can refer my repository. https://github.com/anaghazachariah/video_summary_generaton

[wangsen-sen]

Hello, I want to ask when I want to use generate_ dataset. py to generate my own datasets, always appear “ K = np.dot(video_feat, video_feat.T) #dot是为了获取两个数的乘积 AttributeError: 'NoneType' object has no attribute 'T' ” What should I do?