AI Paper Atlas

Research map / Video / Video generation

Long Video Generation with Time-Agnostic VQGAN and Time-Sensitive Transformer

Songwei Ge, Thomas Hayes, Harry Yang, Xi Yin, Guan Pang, David Jacobs, Jia-Bin Huang, Devi Parikh

ECCV 2022 · Video › Video generation · approach: Autoregressive long video generation · 180 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
condition on class labelsTATS-base · UCF-101IS: 57.63 to 79.28Improved
condition on class labelsTATS-base · UCF-101FVD: 420 -> 332Improved
use replicate padding instead of zero padding in VQGAN (ablation)time-agnostic VQGAN · Padding type study (UCF-101 VQGAN)equivariance score: 0.75Improved
use replicate padding instead of reflect or circular padding (ablation)time-agnostic VQGAN · Padding type study (UCF-101 VQGAN)equivariance scoreImproved
remove all paddings and add real frames to the input (ablation)VQGAN · Padding type study (UCF-101 VQGAN)equivariance score: equivariance score = 1Improved
replace zero padding with replicate padding for sliding-window generation (ablation)vanilla video VQGAN · Sky/Taichi/UCF-101 long videosFVD change over 1024 framesImproved
add hierarchical autoregressive plus interpolation transformer (ablation)TATS-base · UCF-101, Sky, Taichi long videosFVD change over 1024 framesImproved
add hierarchical transformer (ablation)TATS-base · UCF-101 long videosCCS and ICS coherenceImproved
use hierarchical transformer instead of single autoregressive transformerTATS · Generating one 1024-frame videogeneration time: 30 min -> 7.5 minImproved
use TATS instead of DIGAN in human evaluation on SkyTATS · Sky human eval vs DIGANpreference for fewer artifacts: 67%; 355 vs. 175Improved
use TATS-base instead of DIGAN on Sky Time-lapseTATS-base · Sky Time-lapse (16 frames)FVD: 114.6 -> 132.56Hurt
use TATS-base instead of DIGAN on Taichi-HDTATS-base · Taichi-HD (16 frames)FVD: 128.1 -> 94.60Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy