MatReplace is a new reference‑free benchmark for evaluating material replacement in interior scenes, assessing edits on local material correctness, global lighting harmony, outside preservation, and inside structure. It offers three tracks that vary the conditioning signal—instruction only, instruction plus region mask, and material reference image—allowing systematic comparison of different editing approaches. Results show that while closed‑source editors excel at named‑material rendering, grounding materials from pixel references remains a significant challenge.
By Mingzhe Du, Thong Thanh Nguyen, Nguyen Tran Cong Duy, See-Kiong Ng, Luu Anh Tuan
The paper introduces a counterfactual benchmark to evaluate whether vision models understand physical constraints or merely learn rendering shortcuts. Using Mitsuba 3, it generates 5,500 synthetic images across 500 scene families, each with a clean image and ten violations (shadows, reflections, support, surface response, occlusion). Diagnostic baselines (LLaVA‑1.5‑7B, Qwen2.5‑VL‑7B, InternVL3.5‑8B) achieve 64.0–67.8 % macro‑F1 on a held‑out synthetic test set, but performance drops to 40.8 % for LLaVA‑1.5‑7B under an intervention shift, indicating reliance on rendering cues rather than true physical reasoning.
By M. Moein Esfahani, Sepehr Salem, Mohammed Alser, Vince Calhoun
arXiv:2608. 06075v1 Announce Type: cross Abstract: Commercial vision-language models are reshaping computer vision, with visual priors broad enough to rival task-specific systems.
By Shilin Hu, Jingyi Xu, Dimitris Samaras, Hieu Le
Commercial vision-language models are reshaping computer vision, with visual priors broad enough to rival task-specific systems. This raises a natural question: do they reduce the need for classic, physics-informed low-level vision?
SREdit is a response‑guided framework that improves projection‑based appearance editing by estimating a scene response map from a reconstructed scene. This map indicates how strongly projected illumination can affect different surface regions, allowing the system to adapt editing strength across the object while preserving registration with the fixed physical surface. Experiments with two pretrained image editors show that SREdit enhances projected appearance quality and reproduction fidelity, adding less than one second per image with negligible VRAM overhead, and participants consistently preferred its results in a user study.
By Yuqiao Xia, Yuchen Deng, Ziheng Chang, Bingyao Huang
arXiv:2607. 22705v1 Announce Type: cross Abstract: Object-centric learning aims to represent scenes as objects whose properties can be reused in new combinations.
By Anuraag Gadehothur Karnam, Tarunesh Sathish
arXiv:2606. 00188v1 Announce Type: cross Abstract: While current multimodal models are proficient at open-ended visual editing, executing precise single-answer edits remains an important obstacle.
By Kai Xu, Ellis Brown, Shrikar Madhu, Rob Fergus, He He, Saining Xie
arXiv:2610.09125v1 Announce Type: new
Abstract: Reference-based object compositing inserts or replaces an object using a background image, a reference image, and a 2D compositing mask. These inputs g...
By Sanghyun Jo, Chae Yeon Lim, Donghwan Lee, Sihyun Kim, Soo Ye Kim, Kyungsu Kim
STAGE is a training‑free, closed‑form framework for concept erasure in native text‑to‑3D generators. It treats erasure as a stage‑aware editing problem, applying low‑dimensional affine corrections separately to the structural and appearance stages of the pipeline. Experiments on the TRELLIS generator show that STAGE outperforms adapted baselines, achieving a composite score of 66.7 versus 53.2 across 15 shape, material, and object concepts.
By Karol Dziekan, Przemys{\l}aw Spurek, Dawid Malarz
arXiv:2606.13345v2 Announce Type: replace
Abstract: Existing 3D scene editing methods typically rely on per-scene optimization over explicit 3D representations or cascaded edit-and-reconstruct pipeli...
By Xinnan Zhu, Ruijie Xu, Jiayu Ying, Daoguo Dong, Jiachen Xu, Yuan Xie, Xin Tan
arXiv:2606.26738v3 Announce Type: replace
Abstract: While recent advancements in generative image editing models have achieved stunning visual fidelity, it remains an open question whether these syst...
By Tim K\"uchler, Johann-Friedrich Feiden, Matthias Nie{\ss}ner, Carsten Rother
Code4Scene is a benchmark that evaluates coding agents on constructing and editing 3D scenes in Unreal Engine. It tests agents on two tasks: construction, where they must build a scene from open‑ended language, and editing, where they must recover a target scene from reference images while preserving everything else. The benchmark measures task fulfillment, artifact integrity, and physical validity, revealing that construction and editing performance are correlated but not interchangeable, with agents struggling most with spatial composition and precise edits.
By Xiaokang Ye, Siddhant Hitesh Mantri, Zimeng Chen, Edward Zhang, Zhaoxu Zheng, Yuanheng Li, Yizhao Chen, Tianyang Huang, Lianhui Qin