Pith. sign in

REVIEW 1 cited by

End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2311.17241 v2 pith:2MOB6EC6 submitted 2023-11-28 cs.CV

classification cs.CV
keywords end-to-endtrainingbackbonedetectionframesmemoryparametersperformance
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recently, temporal action detection (TAD) has seen significant performance improvement with end-to-end training. However, due to the memory bottleneck, only models with limited scales and limited data volumes can afford end-to-end training, which inevitably restricts TAD performance. In this paper, we reduce the memory consumption for end-to-end training, and manage to scale up the TAD backbone to 1 billion parameters and the input video to 1,536 frames, leading to significant detection performance. The key to our approach lies in our proposed temporal-informative adapter (TIA), which is a novel lightweight module that reduces training memory. Using TIA, we free the humongous backbone from learning to adapt to the TAD task by only updating the parameters in TIA. TIA also leads to better TAD representation by temporally aggregating context from adjacent frames throughout the backbone. We evaluate our model across four representative datasets. Owing to our efficient design, we are able to train end-to-end on VideoMAEv2-giant and achieve 75.4% mAP on THUMOS14, being the first end-to-end model to outperform the best feature-based methods. Code is available at https://github.com/sming256/AdaTAD.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. ProTAL: A Drag-and-Link Video Programming Framework for Temporal Action Localization

    cs.HC 2025-05 conditional novelty 6.0 of 10

    A drag-and-link interface lets users define rules for actions from body-part and object relations, generating frame labels to train temporal action localization models.

Pith tools