AI Paper Atlas

Research map / Video / Video segmentation and tracking

TimeReplayer: Unlocking the Potential of Event Cameras for Video Interpolation

Weihua He, Kaichao You, Zhendong Qiao, Xu Jia, Ziyang Zhang, Wenhui Wang, Huchuan Lu, Yaoyuan Wang, Jianxing Liao

CVPR 2022 · Video › Video segmentation and tracking · approach: Event-based video reconstruction · 42 citations (OpenAlex)

Check 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
Use event streams with frames and unsupervised cycle-consistent training instead of frame-only unsupervised interpolationTimeReplayer · GoPro 7 skip (whole)PSNR: 28.23 -> 34.02Improved
Use event streams with frames and unsupervised cycle-consistent training instead of frame-only unsupervised interpolationTimeReplayer · Middlebury (other) 3 skipPSNR: 24.86 -> 30.91Improved
Use event streams with frames and unsupervised cycle-consistent training instead of frame-only unsupervised interpolationTimeReplayer · Vimeo90k (interpolation) 1 skipPSNR: 30.38 -> 35.12Improved
Use TimeReplayer rather than Time LensVideo interpolation with real events · HQF 1 skipPSNR: 32.49 -> 31.07Hurt
Use TimeReplayer rather than Time LensVideo interpolation with real events · HS-ERGB far 7 skipPSNR: 32.31 -> 30.07Hurt
Use TimeReplayer rather than Time LensVideo interpolation with real events · HS-ERGB close 5 skipPSNR: 32.19 -> 31.21Hurt
Use separate forward and backward event flow estimators instead of a shared estimator (ablation)TimeReplayer · GoPro 7 skipPSNR: 33.78 -> 34.02Improved
Estimate intermediate flows directly rather than proportionally from input-frame flow (ablation)TimeReplayer · GoPro 7 skipPSNR: 32.05 -> 34.02Improved
Use the proposed event-based model rather than UnSuperSloMo with concatenated event input (ablation)TimeReplayer · GoPro 7 skipPSNR: 32.16 -> 34.02Improved
Fine-tune on additional unsupervised DDD17 event dataTimeReplayer · HQF 3 skipPSNR: +2.72Improved
Fine-tune on additional unsupervised DDD17 event dataTimeReplayer · HQF 1 skipPSNR: +2.86Improved

Later papers that compared with this method

1 later accepted paper reports TimeReplayer as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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