Papers › TSM: Temporal Shift Module for Efficient Video Understanding

TSM: Temporal Shift Module for Efficient Video Understanding

20 Nov 2018ICCV 2019 10arXiv:1811.08383archive 2025-07-28

Ji Lin, Chuang Gan, Song Han

The explosive growth in video streaming gives rise to challenges on performing video understanding at high accuracy and low computation cost. Conventional 2D CNNs are computationally cheap but cannot capture temporal relationships; 3D CNN based methods can achieve good performance but are computationally intensive, making it expensive to deploy. In this paper, we propose a generic and effective Temporal Shift Module (TSM) that enjoys both high efficiency and high performance. Specifically, it can achieve the performance of 3D CNN but maintain 2D CNN's complexity. TSM shifts part of the channels along the temporal dimension; thus facilitate information exchanged among neighboring frames. It can be inserted into 2D CNNs to achieve temporal modeling at zero computation and zero parameters. We also extended TSM to online setting, which enables real-time low-latency online video recognition and video object detection. TSM is accurate and efficient: it ranks the first place on the Something-Something leaderboard upon publication; on Jetson Nano and Galaxy Note8, it achieves a low latency of 13ms and 35ms for online video recognition. The code is available at: https://github.com/mit-han-lab/temporal-shift-module.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1811.08383")

Code

Syntology Ran 6 of 16 code samples harvested from 2 repositories linked to this paper; 10 have no recorded run. Of those that ran: 1 ran · honoured contract; 5 ran · our draft was wrong.

By repository: community (archive-listed): 13 samples from 2 repositories, 4 ran; 3 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

13 repositories listed; official and paper-mentioned ones first.

MIT-HAN-LAB/temporal-shift-module officialmentioned in papermentioned on GitHubpytorchMIT report
PaParaZz1/TemporalShiftModule mentioned on GitHubpytorch report
WavesUR/embedded_TSM mentioned on GitHubpytorch report
bespontaneous/ffn mentioned on GitHubpytorchMIT report
bespontaneous/mca-pytorch mentioned on GitHubpytorch report
rijuldhir/TSM mentioned on GitHubpytorch report
stevedepp/yolo4asl mentioned on GitHub report
sunutf/TSM mentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

16 samples harvested; 6 ran; 1 honoured the contract we drafted; 10 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
5ran · our draft was wrong
10unverified

Licence: 4 of the 16 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

conv1x1x1 bespontaneous/mca-pytorch/2D_Network/ops/backbone/resnet_TSM.py community (archive-listed) ran · our draft was wrong no licence file found · pointer only · dfcfb80c1514fe42 · report
conv_1x1_bn sunutf/TSM/online_demo/mobilenet_v2_tsm.py community (archive-listed) ran · our draft was wrong MIT (permissive) · a0131fb70c267a9e · report
conv_bn sunutf/TSM/online_demo/mobilenet_v2_tsm.py community (archive-listed) ran · our draft was wrong MIT (permissive) · e087cbfa1e53486d · report
make_divisible sunutf/TSM/online_demo/mobilenet_v2_tsm.py community (archive-listed) ran · honoured contract fingerprinted MIT (permissive) · 63ec083689f80f47 · report
bninception sunutf/TSM/archs/bn_inception.py community (archive-listed) unverified MIT (permissive) · c89c29bd79c97227 · report
build_dataflow sunutf/TSM/main_amp.py community (archive-listed) unverified MIT (permissive) · dc07c7d74de65f4f · report
init_hidden sunutf/TSM/ops/models_rnn.py community (archive-listed) unverified MIT (permissive) · 05f08d45ec764f25 · report
input_fn sunutf/TSM/tsm_fpga/fpga_build/model_tf_split/calib_input_split.py community (archive-listed) unverified MIT (permissive) · a866d9f9275c700a · report
process_output sunutf/TSM/tsm_fpga/main_tf.py community (archive-listed) unverified MIT (permissive) · 0be9a22aa03cb8bb · report
torch_params sunutf/TSM/tsm_fpga/mobilenet_v2_tfslim.py community (archive-listed) unverified MIT (permissive) · 47e316f6d90b3490 · report
torch_to_conv_state sunutf/TSM/tsm_fpga/mobilenet_v2_tfslim.py community (archive-listed) unverified MIT (permissive) · 9823b80aaa256ebd · report
torch_to_linear_state sunutf/TSM/tsm_fpga/mobilenet_v2_tfslim.py community (archive-listed) unverified MIT (permissive) · b85fde19d195b87f · report
transform sunutf/TSM/tsm_fpga/main_tf.py community (archive-listed) unverified MIT (permissive) · 406596818e861a3d · report
conv1x1 identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · d9def42110729a85 · report
conv3x3 identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · fac5364e2f53c6db · report
make_temporal_shift identical code first harvested elsewhere unverified licence of this copy not recorded · ea903b20756dc01d · report

Tasks

3D Action RecognitionAction ClassificationAction RecognitionMicro-Action RecognitionObject DetectionVideo Object DetectionVideo Object TrackingVideo RecognitionVideo Understandingobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
3D Action Recognition Assembly101 TSM Actions Top-1 35.27 #2 of 7 Archive leaderboard report
3D Action Recognition Assembly101 TSM Object Top-1 47.45 #2 of 7 Archive leaderboard report
3D Action Recognition Assembly101 TSM Verbs Top-1 58.27 #2 of 7 Archive leaderboard report
Action Classification Kinetics-400 TSM Acc@1 74.7 #161 of 207 Archive leaderboard report
Action Recognition Something-Something V1 TSM (RGB + Flow) Top 1 Accuracy 50.7 #50 of 74 Archive leaderboard report
Action Recognition Something-Something V1 TSMEn Top 1 Accuracy 49.7 #54 of 74 Archive leaderboard report
Action Recognition Something-Something V1 TSMEn Top 5 Accuracy 78.5 #54 of 74 Archive leaderboard report
Action Recognition Something-Something V1 TSM Top 1 Accuracy 47.2 #65 of 74 Archive leaderboard report
Action Recognition Something-Something V1 TSM Top 5 Accuracy 77.1 #65 of 74 Archive leaderboard report
Action Recognition Something-Something V2 TSM (RGB + Flow) Top-1 Accuracy 66.6 #76 of 123 Archive leaderboard report
Action Recognition Something-Something V2 TSM (RGB + Flow) Top-5 Accuracy 91.3 #76 of 123 Archive leaderboard report
Video Object Detection ImageNet VID Online TSM MAP 76.3 #30 of 33 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections