SlowFast Networks for Video Recognition
Christoph Feichtenhofer, Haoqi Fan, Jitendra Malik, Kaiming He
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/facebookresearch/SlowFastOfficialIn paperpytorch★ 7,320
- github.com/open-mmlab/mmaction2pytorch★ 4,959
- github.com/facebookresearch/pytorchvideopytorch★ 3,551
- github.com/wangxiang1230/SSTAPpytorch★ 72
- github.com/tianfr/mononerfpytorch★ 70
- github.com/LukasHedegaard/co3dpytorch★ 45
- github.com/tonysy/PyActionpytorch★ 17
- github.com/tianfr/semantic-flowpytorch★ 13
- github.com/Ysp9714/SlowFastNet-kerastf★ 0
- github.com/PaddlePaddle/PaddleVideo/blob/develop/docs/zh-CN/model_zoo/recognition/slowfast.mdpaddle★ 0
Abstract
We present SlowFast networks for video recognition. Our model involves (i) a Slow pathway, operating at low frame rate, to capture spatial semantics, and (ii) a Fast pathway, operating at high frame rate, to capture motion at fine temporal resolution. The Fast pathway can be made very lightweight by reducing its channel capacity, yet can learn useful temporal information for video recognition. Our models achieve strong performance for both action classification and detection in video, and large improvements are pin-pointed as contributions by our SlowFast concept. We report state-of-the-art accuracy on major video recognition benchmarks, Kinetics, Charades and AVA. Code has been made available at: https://github.com/facebookresearch/SlowFast
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| AVA v2.1 | SlowFast (Kinetics-400 pretraining) | mAP (Val) | 26.3 | — | Unverified |
| AVA v2.1 | SlowFast++ (Kinetics-600 pretraining, NL) | mAP (Val) | 28.3 | — | Unverified |
| AVA v2.1 | SlowFast (Kinetics-600 pretraining, NL) | mAP (Val) | 27.3 | — | Unverified |
| AVA v2.1 | SlowFast (Kinetics-600 pretraining) | mAP (Val) | 26.8 | — | Unverified |
| AVA v2.2 | SlowFast, 16x8 R101+NL (Kinetics-600 pretraining) | mAP | 27.5 | — | Unverified |
| AVA v2.2 | SlowFast, 4x16, R50 (Kinetics-400 pretraining) | mAP | 21.9 | — | Unverified |
| AVA v2.2 | SlowFast, 8x8, R101 (Kinetics-400 pretraining) | mAP | 23.8 | — | Unverified |
| AVA v2.2 | SlowFast, 8x8 R101+NL (Kinetics-600 pretraining) | mAP | 27.1 | — | Unverified |
| Diving-48 | SlowFast | Accuracy | 77.6 | — | Unverified |
| H2O (2 Hands and Objects) | SlowFast | Actions Top-1 | 77.69 | — | Unverified |
| Something-Something V2 | SlowFast | Top-1 Accuracy | 61.7 | — | Unverified |