AI Paper Atlas

Research map / Video / Video segmentation and tracking

Referring Segmentation in Images and Videos with Cross-Modal Self-Attention Network

Linwei Ye, Mrigank Rochan, Zhi Liu, Xiaoqin Zhang, Yang Wang

TPAMI 2021 · Video › Video segmentation and tracking · approach: Video semantic and panoptic segmentation · 94 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 word attention-aware pooling instead of max pooling (ablation)CMSA · G-Ref valOverall IoU: 32.96 -> 37.35Improved
use word attention-aware pooling instead of average pooling (ablation)CMSA · G-Ref valOverall IoU: 33.41 -> 37.35Improved
use word attention-aware pooling instead of average pooling (ablation)CMSA · UNC+ valOverall IoU: 32.14 -> 34.60Improved
use cross-modal self-attention instead of no attention (ablation)CMSA · UNC valOverall IoU: 45.63 -> 50.95Improved
use cross-modal self-attention instead of word attention only (ablation)CMSA · UNC valOverall IoU: 47.01 -> 50.95Improved
use cross-modal self-attention instead of pixel attention only (ablation)CMSA · UNC valOverall IoU: 47.84 -> 50.95Improved
use cross-modal self-attention instead of word-pixel pair attention (ablation)CMSA · UNC valOverall IoU: 47.57 -> 50.95Improved
keep per-word vectors (CMSA-W) instead of one LSTM sentence vector (CMSA-S) (ablation)CMSA · UNC valOverall IoU: 48.53 -> 50.95Improved
use gated multi-level fusion (GMLF) instead of pyramid pooling module (ablation)CMSA · UNC valOverall IoU: 53.54 -> 58.92Improved
use gated multi-level fusion (GMLF) instead of ConvLSTM (ablation)CMSA · UNC valOverall IoU: 56.56 -> 58.92Improved
use cross-level gated fusion instead of self-gated fusion (ablation)CMSA+GMLF · UNC valOverall IoU: 57.08 -> 58.92Improved
add cross-frame self-attention (CFSA)Ours · A2D SentencesOverall IoU: 59.2 -> 61.8Improved

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