CitationWhen using the DET or CLS-LOC dataset, please cite:¬
Olga Russakovsky*, Jia Deng*, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg and Li Fei-Fei. (* = equal contribution) ImageNet Large Scale Visual Recognition Challenge. arXiv:1409.0575, 2014.
Bolei Zhou, Aditya Khosla, Agata Lapedriza, Antonio Torralba and Aude Oliva. Places2: A Large-scale Database for Scene Understanding. Arxiv, 2015. (coming soon)
Object detection from video (VID)
- Overview and statistics of the data.
- Meta data for the competition categories.
- Matlab routines for evaluating submissions.
Please be sure to consult the included readme.txt file for VID competition details. Additionally, the development kit includes
VID initial release. 22GB. MD5: b329300dd0cd4422171878970d30e1da
VID initial release snippets. 15GB. MD5: 4e8f46f7d507edec5a42e1c25de664c3
There are a total of 1952 snippets for training. The number of snippets for each synset ranges from 23 to 261. There are 281 validation snippets and 458 test snippets. All snippets are extracted into frames in JPEG format. Original snippets are also provided.
VID final releasenew 43GB. MD5: ec74f41fc65873eaa55abafc75db23b4
VID final release snippetsnew 7.8GB. MD5: f338ddd0765e0703097b7462a4fb3186
These are the new data. There are a total of 1910 snippets for training. There are 274 validation snippets and 479 test snippets. All snippets are extracted into frames in JPEG format. Original snippets are also provided.
VIDnew 86GB. MD5: 5feb88ac3345e5b5eae71f6ec8a91325
These are the whole data, including everything that is needed for the task. There are a total of 3862 snippets for training. The number of snippets for each synset (category) ranges from 56 to 458. There are 555 validation snippets and 937 test snippets. All snippets are extracted into frames in JPEG format. Original snippets are also provided.