English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We propose a new formulation of temporal action detection (TAD) withdenoising diffusion, DiffTAD in short. Taking as input random temporalproposals, it can yield action proposals accurately given an untrimmed longvideo. This presents a generative modeling perspective, against previousdiscriminative learning manners. This capability is achieved by first diffusingthe ground-truth proposals to random ones (i.e., the forward/noising process)and then learning to reverse the noising process (i.e., the backward/denoisingprocess). Concretely, we establish the denoising process in the Transformerdecoder (e.g., DETR) by introducing a temporal location query design withfaster convergence in training. We further propose a cross-step selectiveconditioning algorithm for inference acceleration. Extensive evaluations onActivityNet and THUMOS show that our DiffTAD achieves top performance comparedto previous art alternatives. The code will be made available athttps://github.com/sauradip/DiffusionTAD.

DiffTAD: Temporal Action Detection with Proposal Denoising Diffusion