facebookresearch/SlowFast

View on GitHub

PySlowFast: video understanding codebase from FAIR for reproducing state-of-the-art video models.

Stars: 7.4k
Forks: 1.3k
Open beginner issues: 0
Indexed issues: 424

Dominant language: Python
License: Apache-2.0
Last GitHub push: Mar 16, 2026
Latest indexed: Sep 15, 2026
Contributing guide: Contributing guide
Code of conduct: Code of conduct
Beginner labels: good first issue help wanted
PR merge metrics: No merged PRs in 30d

Open All

424 open issues indexed

How to fine-tune SlowFast Kinetics model Open

enhancement

facebookresearch/SlowFast#18 · 104 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Charades Model Open

enhancement

facebookresearch/SlowFast#26 · 0 comments · 17 reactions · 0 assignees · Indexed Sep 15, 2026
Demo programme on camera input Open

enhancement good first issue

facebookresearch/SlowFast#34 · 57 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
filter visualization Open

enhancement

facebookresearch/SlowFast#36 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Is it possible to just use pre-trained model without GPU Open

enhancement

facebookresearch/SlowFast#39 · 6 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
How is this demo "ava_demo.gif" generated? Open

enhancement

facebookresearch/SlowFast#51 · 8 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
model zool link erro. Open

enhancement

facebookresearch/SlowFast#56 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Lower number of residual blocks uses more GPU memory Open

question

facebookresearch/SlowFast#63 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Gray-scale network and pre-trained model? Open

enhancement

facebookresearch/SlowFast#68 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Running demo on person scale Open

enhancement

facebookresearch/SlowFast#72 · 0 comments · 5 reactions · 0 assignees · Indexed Sep 15, 2026
Training SLOWFAST_16x8_R101_50_50 Open

enhancement

facebookresearch/SlowFast#74 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Questions about gpu memory of slowfast network and I3D. Open

question

facebookresearch/SlowFast#78 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
When to use SyncBatchnorm Open

enhancement

facebookresearch/SlowFast#79 · 1 comment · 1 reaction · 0 assignees · Indexed Sep 15, 2026
Could you release the Kinetics-600 pretrained models of SlowFast? Open

enhancement

facebookresearch/SlowFast#82 · 0 comments · 6 reactions · 0 assignees · Indexed Sep 15, 2026
lake8 testing SlowFast on Python 3.8.0 Open

enhancement

facebookresearch/SlowFast#84 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Could you release the classes id for kinetics 600 pretrained models? Open

enhancement

facebookresearch/SlowFast#95 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Possibility to perform action localization? Open

enhancement

facebookresearch/SlowFast#99 · 1 comment · 1 reaction · 0 assignees · Indexed Sep 15, 2026
pretrained model performed very bad on K400 Open

question

facebookresearch/SlowFast#100 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Run on CPU Open

enhancement

facebookresearch/SlowFast#103 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
R101 ImageNet Pretrained model Open

enhancement

facebookresearch/SlowFast#107 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Problem about bounding box size and how to replicate Detection model settings ? Open

documentation

facebookresearch/SlowFast#115 · 0 comments · 3 reactions · 0 assignees · Indexed Sep 15, 2026
Exception: process 0 terminated with signal SIGSEGV Open

invalid

facebookresearch/SlowFast#118 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Release of 32-frame models? (16x8, etc) Open

enhancement

facebookresearch/SlowFast#125 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
print summary Open

enhancement

facebookresearch/SlowFast#126 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 15, 2026
Small error in Kinetics model output head (validation loss only) Open

enhancement

facebookresearch/SlowFast#128 · 0 comments · 3 reactions · 0 assignees · Indexed Sep 15, 2026
AVA results visualization Open

enhancement

facebookresearch/SlowFast#133 · 0 comments · 5 reactions · 0 assignees · Indexed Sep 15, 2026
Activity list for Kinetics 600 Open

documentation

facebookresearch/SlowFast#134 · 5 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Cannot reproduce the result on AVA Open

enhancement

facebookresearch/SlowFast#141 · 22 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
about FC layer output size Open

question

facebookresearch/SlowFast#143 · 2 comments · 2 reactions · 0 assignees · Indexed Sep 15, 2026
Are you planning to release audio visual slowfast? Open

enhancement

facebookresearch/SlowFast#144 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
length of sampled clips indicated in the paper Open

documentation

facebookresearch/SlowFast#146 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Preparation of AVA-Dataset Format for Training on Custom Dataset Open

documentation

facebookresearch/SlowFast#149 · 74 comments · 9 reactions · 0 assignees · Indexed Sep 15, 2026
Testing on custom video Open

enhancement

facebookresearch/SlowFast#152 · 2 comments · 8 reactions · 0 assignees · Indexed Sep 15, 2026
Questions in detector architecture of slow fast for AVA dataset Open

question

facebookresearch/SlowFast#156 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
training loss is very low (0.177622) at the very beginning, is it normal? Open

question

facebookresearch/SlowFast#159 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Training on new dataset Open

question

facebookresearch/SlowFast#162 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Cannot load C2D pre-trained model Open

enhancement

facebookresearch/SlowFast#163 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
AVA video data problem Open

question

facebookresearch/SlowFast#165 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
training on a new dataset Open

question

facebookresearch/SlowFast#166 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
About Audio-Visual Slowfast Open

facebookresearch/SlowFast#167 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
set up error about "av" Open

question

facebookresearch/SlowFast#168 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
is there a little error Open

question

facebookresearch/SlowFast#169 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
action detection train and test Open

question

facebookresearch/SlowFast#171 · 0 comments · 2 reactions · 0 assignees · Indexed Sep 15, 2026
Runtime error Open

help wanted

facebookresearch/SlowFast#172 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
will the code save logs to a file when training? Open

enhancement

facebookresearch/SlowFast#175 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
An error in temporal sampling? Open

question

facebookresearch/SlowFast#176 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
missing of sth-sthv2 frame list Open

enhancement

facebookresearch/SlowFast#177 · 1 comment · 2 reactions · 0 assignees · Indexed Sep 15, 2026
Memory Problem Open

question

facebookresearch/SlowFast#180 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 15, 2026
Torchvision backend not working. Open

question

facebookresearch/SlowFast#181 · 35 comments · 2 reactions · 0 assignees · Indexed Sep 15, 2026
Bug: target_fps in the decoder shouldn't be an integer in NTSC system. Open

enhancement

facebookresearch/SlowFast#182 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 15, 2026

facebookresearch/SlowFast

Get new issues in your inbox