AI Paper Atlas

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

Kiroshi: An Agentic Perception System for High-Accuracy Image Parsing

Haipeng ZHOU, Jinshan Liu, He Zhang, Xuequan Lu, Jun Ma, Lei Zhu

ECCV 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 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
add DPO-aligned MLLM policy with 3 refinement rounds vs vanilla box (ablation)Kiroshi · HiFiRefMS segmentationmaxF_beta: 0.764 -> 0.875Improved
add DPO-aligned MLLM policy with 3 refinement rounds vs vanilla box (ablation)Kiroshi · HiFiRefMS mattingSAD: 32.16 -> 26.34Improved
increase refinement rounds from 3 to 5 (ablation)Kiroshi · HiFiRefMS segmentationmaxF_beta: 0.875 -> 0.863Hurt
increase refinement rounds from 1 to 3 (ablation)Kiroshi · HiFiRefMS segmentationHCE: 1514 -> 1218Improved
use 32x32 grid instead of 64x64 (ablation)Kiroshi · HiFiRefMS segmentationmaxF_beta: 0.820 -> 0.875Improved
use 32x32 grid instead of 16x16 (ablation)Kiroshi · HiFiRefMS segmentationmaxF_beta: 0.862 -> 0.875Improved
use Ovis2.5-2B instead of Qwen2.5-VL-3B backbone (ablation)Kiroshi · HiFiRefMS segmentationmaxF_beta: 0.852 -> 0.875Improved
use Qwen3-VL-2B instead of Ovis2.5-2B backbone (ablation)Kiroshi · HiFiRefMS mattingSAD: 26.34 -> 24.37Mixed
use Kiroshi instead of ZIMKiroshi · HiFiRefMS mattingSAD: 36.41 -> 26.34Improved
run mask-guided refinement x3 with external pipelines instead of x1Text4Seg + MGMatting · HiFiRefMS segmentationmaxF_beta: 0.846 -> 0.833Hurt
run Kiroshi refinement x3 instead of x1Kiroshi · HiFiRefMS segmentationmaxF_beta: 0.838 -> 0.875Improved

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