AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

You Only Infer Once: Cross-Modal Meta-Transfer for Referring Video Object Segmentation

Dezhuang Li, Ruoqi Li, Lijun Wang, Yifan Wang, Jinqing Qi, Lu Zhang, Ting Liu, Qingquan Xu, Huchuan Lu

AAAI 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 47 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
use one-stage YOFO instead of URVOSRVOS · Refer-DAVIS2017 validation, first-frame expressionJ &F: 51.63 -> 54.44Improved
use YOFO instead of RefVOSRVOS · Refer-DAVIS2017 validation, full-video expressionJ &F: 45.1 -> 52.16Improved
use YOFO instead of CMPC-VRVOS · Refer-Youtube-VOS validationJ &F: 47.48 -> 48.59Improved
use YOFO instead of URVOSRVOS · Refer-Youtube-VOS validationJ &F: 47.23 -> 48.59Improved
use multi-scale cross-modal feature mining instead of the baseline fusion scheme (ablation)YOFO · Refer-DAVIS2017 validation, first-frame expressionJ &F: 45.2 -> 49.5Improved
use multi-scale rather than single-scale cross-modal feature mining (ablation)YOFO · Refer-DAVIS2017 validation, first-frame expressionJ &F: 43.5 -> 49.5Improved
use the proposed feature mining rather than concatenating image and language features at each scale (ablation)YOFO · Refer-DAVIS2017 validation, first-frame expressionJ &F: 42.9 -> 49.5Improved
add the Meta-Transfer module (ablation)YOFO · Refer-DAVIS2017 validation, first-frame expressionJ &F: 42.7 -> 49.5Improved
add augmented memory of recent frames for meta-transfer learning (ablation)YOFO · Refer-DAVIS2017 validation, first-frame expressionJ &F: 48.4 -> 49.5Improved

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