AI Paper Atlas

Research map / Video / Video segmentation and tracking

Semantic-Aware Dynamic Parameter for Video Inpainting Transformer

Eunhye Lee, Jinsu Yoo, Yunjeong Yang, Sungyong Baik, Tae Hyun Kim

ICCV 2023 · Video › Video segmentation and tracking · approach: Video semantic and panoptic segmentation · 5 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 super-categorized segmentation map as input (ablation)FuseFormer-small · DAVISPSNR: 30.51 -> 30.90Improved
add semantic-aware dynamic linear layer naively (ablation)FuseFormer-small · DAVISPSNR: 30.90 -> 30.92No clear effect
add token-wise classification loss (ablation)SAVIT · DAVISPSNR: 30.92 -> 31.19Improved
add semantic-aware discriminator (ablation)SAVIT · DAVISPSNR / VFID: PSNR 31.19 -> 31.01; VFID 0.165 -> 0.157Mixed
use hardmax instead of softmax in router (ablation)SAVIT · DAVISPSNR: 30.91 -> 31.01Improved
replace trained routing weights with random weights (ablation)SAVIT · DAVISPSNR: 31.01 -> 19.86Hurt
increase number of experts from 4 to 8 (ablation)SAVIT · DAVISPSNR: 30.83 -> 31.01Improved
increase number of semantic blocks from 1 to 2 (ablation)SAVIT · DAVISPSNR: 30.40 -> 30.45Improved
add semantic-aware module under large masksFuseFormer-small · DAVIS mask ratio 60%PSNR: 18.10 -> 19.21Improved

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