AI Paper Atlas

Research map / 3D vision / 3D generation

Segment Anything in 3D with Radiance Fields

Jiazhong Cen, Jiemin Fang, Zanwei Zhou, Chen Yang, Lingxi Xie, Xiaopeng Zhang, Wei Shen, Qi Chuan Tian

IJCV 2025 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 110 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 3D distance based confidence decay to self-prompting (ablation)SA3D-TensoRF · NVOSmIoU (%): 87.9 -> 90.3Improved
add confidence decay on hard scene LLFF-trex (ablation)SA3D-TensoRF · NVOS TrexIoU (%): 64.0 -> 82.0Improved
add confidence decay on easy scenes (ablation)SA3D-TensoRF · NVOS other scenesIoU (%): Fern 82.9 -> 82.9No clear effect
use SAM instead of SEEM as 2D segmenter (ablation)SA3D-TensoRF · NVOSmIoU (%): 86.0 -> 90.3Improved
use SAM instead of RITM as 2D segmenter (ablation)SA3D-TensoRF · NVOSmIoU (%): 81.2 -> 90.3Improved
increase mask grid resolution from 40^3 to 320^3 (ablation)SA3D-TensoRF · NVOSmIoU (%): 88.0 -> 90.3Improved
use explicit mask grids instead of an MLP mask (ablation)SA3D-TensoRF · NVOSmIoU (%): 80.9 -> 90.3Improved
add feature cache to SA3D-TensoRFSA3D-TensoRF · NVOSmIoU (%) / time: 90.3 -> 90.1 mIoU; 45 s -> 30 s + 15Mixed
add feature cache to SA3D-GSSA3D-GS · Horns self-prompting timems per step: 508.1 -> 40.7Improved
replace TensoRF/NeRF with 3D-GS as radiance fieldSA3D · NVOSmIoU (%): 90.1 -> 92.2Improved
replace TensoRF with 3D-GS as radiance fieldSA3D · Replica meanmIoU (%): 82.7 -> 79.1Hurt
add iterative multi-view SA3D over single-view mask inversionSA3D-GS · SPIn-NeRF meanIoU (%): 74.6 -> 93.2Improved

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