AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

SAM4MLLM: Enhance Multi-Modal Large Language Model for Referring Expression Segmentation

Yi-Chia Chen, Wei-Hua Li, Cheng Sun, Yu-Chiang Frank Wang, Chu-Song Chen

ECCV 2024 · Vision-language and multimodal › Human and embodied modeling · approach: Referring expression comprehension · 53 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 PQPP (proactive point query) instead of PPG (direct point generation) (ablation)SAM4MLLM Qwen-VL-7B · refCOCOg val(U)cIoU: 74.2 -> 74.5Improved
use PQPP instead of PPG (ablation)SAM4MLLM Qwen-VL-7B · refCOCO valcIoU: 76.2 -> 77.1Improved
replace Qwen-VL backbone with LLaVA1.6 for PQPP (ablation)SAM4MLLM-PQPP · refCOCO valcIoU: 77.1 -> 79.6Improved
set point confidence threshold to 0.9 versus 0.6 (ablation)SAM4MLLM-PQPP · RefCOCOg valcIoU: 69.5 -> 74.5Improved
raise threshold from 0.9 to 0.95 (ablation)SAM4MLLM-PQPP · RefCOCOg valcIoU: 74.5 -> 73.8Hurt
use 5x5 grid instead of random sampling of 25 points (ablation)SAM4MLLM-PQPP · RefCOCOg valcIoU: 73.7 -> 74.5Improved
use 6x6 grid instead of 5x5 grid (ablation)SAM4MLLM-PQPP · RefCOCOg valcIoU: 74.5 -> 74.4No clear effect
add more mask training data (SAM4MLLM*) with LLaVA1.6 backboneSAM4MLLM-8B-PQPP · ReasonSeg val (zero-shot)gIoU: 46.7 -> 58.4Improved
fine-tune MLLM with SAM prompt dataQwen-VL-Chat-7B · VQAv2accuracy: 78.2 -> 78.7Improved
use SAM4MLLM instead of GLaMM on simple short queriesrefCOCO / refCOCO+ vs refCOCOg · refCOCO val vs refCOCOg valcIoU: refCOCO 77.1 vs 79.5; refCOCOg 74.5 vs 74.2Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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