AI Paper Atlas

Research map / Video / Video segmentation and tracking

Video Frame Synthesis Using Deep Voxel Flow

Ziwei Liu, Raymond A. Yeh, Xiaoou Tang, Yiming Liu, Aseem Agarwala

ICCV 2017 · Video › Video segmentation and tracking · approach: Video frame interpolation · 789 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
add multi-scale voxel flow to DVF (ablation)Deep Voxel Flow · UCF-101 test setPSNRImproved
add skip connections to the network (ablation)Deep Voxel Flow · UCF-101PSNR: 1.1dBImproved
fine-tune on KITTI training setDeep Voxel Flow · KITTI view synthesisL1 error: 0.336 -> 0.178Improved
fine-tune on KITTI before evaluating flowDeep Voxel Flow · KITTI flow 2012EPE: 14.6 -> 9.5Improved
fine-tune with action recognition loss after frame-synthesis pretrainingDeep Voxel Flow · UCF-101 action recognitionaccuracy: 48.7 -> 52.4Improved
use DVF frame-synthesis pretraining instead of random initializationaction recognition network · UCF-101 action recognitionaccuracy: 39.1 -> 52.4Improved
replace Beyond MSE hallucination with optical-flow pixel copyingvideo interpolation · UCF-101 interpolationPSNR: 1.4dBImproved
replace EpicFlow with learned voxel flowvideo interpolation · UCF-101 interpolationPSNR: 1.6dBImproved
replace EpicFlow with learned voxel flowvideo extrapolation · UCF-101 extrapolation (2 frames)PSNR: 31.3 -> 32.7Improved
use 4 input frames instead of 2 for extrapolationDeep Voxel Flow · UCF-101 extrapolationPSNR: 32.7 -> 33.4Improved
apply UCF-101-trained DVF directly to KITTI view synthesis versus prior methodsDeep Voxel Flow · KITTI view synthesisL1 error: 0.164 and 0.135Improved
use unsupervised DVF flow instead of EpicFlow for flow estimationflow estimation · KITTI flow 2012EPE: 3.8 -> 9.5Hurt

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