AI Paper Atlas

Research map / Detection and segmentation / CNN features

ET-SAM: Efficient Point Prompt Prediction in SAM for Unified Scene Text Detection and Layout Analysis

Xike Zhang, Maoyuan Ye, Juhua Liu, Bo Du

ECCV 2026 · Detection and segmentation › CNN features · approach: Arbitrary-shape scene text detection · 0 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
replace Hi-SAM's S-Decoder random 1500 points with the lightweight point decoder (ablation)Hi-SAM-L · HierText testaverage time: 4.4s -> 1.2s with PQ 63.10 -> 63.50 (word)Improved
reduce Hi-SAM random sampled points from 1500 to 280 (ablation)Hi-SAM-L · HierText testword PQ: 63.10 -> 61.04Hurt
add point and mask task prompts together (ablation)ET-SAM · Total-TextF-score: 81.99 -> 86.37Improved
add point and mask task prompts together (ablation)ET-SAM · HierText layout analysisF-score: 73.38 -> 74.51Improved
add a single task prompt alone (ablation)ET-SAM · CTW1500F-score: 86.79 -> 86.30 (mask only), 86.75 (point only), 87.14 (both)Mixed
add Total-Text and CTW1500 to joint training data (ablation)ET-SAM · Total-Text and CTW1500F-score: Total-Text 69.02 -> 85.93Improved
add Total-Text and CTW1500 to joint training data (ablation)ET-SAM · HierText testword PQ: 63.50 -> 61.96Hurt
raise point threshold to 0.8 (ablation)ET-SAM · HierText testword PQ: 62.52 -> 56.98Hurt
change heatmap loss weight away from 50 (ablation)ET-SAM · HierText testword PQ: 62.52 -> 61.61 (25), 62.06 (100)Hurt
fine-tune after joint trainingET-SAM · HierText testword PQ: 62.52 -> 63.60Improved
ET-SAM vs Hi-SAM-L in joint trainingET-SAM · Total-Text, CTW1500, ICDAR2015F-score: 70.2 -> 87.0 (Total-Text); 77.4 -> 87.9 (CTW1500); 77.4 -> 83.2 (ICDAR2015)Improved
ET-SAM vs specialist detectorsET-SAM · Total-TextF-score: 87.0 vs 90.0 (DeepSolo)Hurt

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