CogCanvas is a new benchmark for multi-subject reference-based image generation, featuring 1,952 curated reference images of 100 celebrities, 115 objects/fashion items, and 29 real-world backgrounds. It generates 1,361 compositional prompts with 2–5 people, using a pipeline that includes DINOv2 deduplication, aesthetic filtering, and automated graph derivation for interaction and positioning. The benchmark evaluates three tasks—reference-based multi-human-object generation, text-to-image compositional generation, and reference retrieval—under a six-axis protocol, and introduces BG‑Sim and Attr‑VQA metrics to assess background fidelity and attribute binding.
By Long-Bao Nguyen, Quang-Khai Le, Tam V. Nguyen, Minh-Triet Tran, Trung-Nghia Le
arXiv:2605. 09233v2 Announce Type: replace-cross Abstract: Recent advances in visual generative models have enabled high-fidelity image editing guided by human instructions.
By Zilai Zeng, Mingdeng Cao, Zijie Li, Xiaochen Lian, Yichun Shi, Peihao Zhu, Chen Sun, Peng Wang
Recent advances in unified multimodal models have significantly improved text-guided image editing abilities. In particular, models such as Nano-Banana-Pro and GPT-Image-2 demonstrate emerging capabilities in multi-source image editing (MIE), including tasks such as object synthesis, person-background composition, and cross-image style fusion.
arXiv:2607. 21318v1 Announce Type: cross Abstract: Replacing an object with one that differs in category or shape requires complete source removal, natural target formation unconstrained by the source silhouette, and preservation of unrelated content.
By Jian Zhang, Zhijun Zhang
arXiv:2606. 08016v1 Announce Type: cross Abstract: Current image editing software often hinges on fixed filters or expert tuning, leaving a gap between amateur users' intent and outcomes.
By Zichen Zhu, Yuheng Sun, Mingxuan Zhu, Wenjie Ma, Situo Zhang, Zhexiang Wang, Ziyue Yang, Danyang Zhang, Kunyao Lan, Zihan Zhao, Dingye Liu, Siqi Xiang, Lu Chen, Kai Yu
Replacing an object with one that differs in category or shape requires complete source removal, natural target formation unconstrained by the source silhouette, and preservation of unrelated content. Existing training-free editors either localize edits from terminal predictions under source and target prompts or preserve unrelated content through spatially unselective source-feature reuse without explicit region discovery.
TransPhy is a framework for visually in-context learning that focuses on physically grounded image editing. It introduces PhysVICL-74, a dataset of 74 transformation rules and 5,240 source–target pairs, and evaluates models on novel-instance transfer and unseen-rule generalization. The method predicts the demonstrated rule and a query-specific target-state description, then synthesizes the target image using token-wise mixture-of-experts guided by localized transition cues, improving rule adherence, query consistency, and generalization over existing methods.
By Siyi Xie, Xuanke Shi, Jinsheng Quan, Haoran Tang, Zukai Chen, Lei Yang, Quan Wang
arXiv:2507. 17853v2 Announce Type: replace-cross Abstract: Recent advances in text-to-image (T2I) generation have led to impressive visual results.
By Lifeng Chen, Jiner Wang, Zihao Pan, Beier Zhu, Xiaofeng Yang, Chi Zhang
The demand for image manipulation has seen a significant increase recently. Traditional tools like Photoshop and Capture One, while powerful, require considerable expertise to use effectively.
arXiv:2510. 08532v2 Announce Type: replace-cross Abstract: Instruction-based image editing offers a powerful and intuitive way to manipulate images through natural language.
By Rishubh Parihar, Or Patashnik, Daniil Ostashev, R. Venkatesh Babu, Daniel Cohen-Or, Kuan-Chieh Wang
arXiv:2607. 07051v1 Announce Type: cross Abstract: Conversational image editing requires preserving not only visible content, but also content that temporarily disappears across turns.
By Soomin Han, Jihyung Ahn, Bumsoo Kim, Buru Chang
The paper introduces the Identity-Aware Human-Object Interaction Motion Captioning task, which requires captions to include both the subject’s identity and the interaction motion, e.g., "Sub_ID lifts the chair" instead of a generic description. It proposes ID‑HOINet, a model that learns from multi‑view videos using a Multi‑View Identity‑Motion Learning Module and a Two‑Stage Caption Rewriting Strategy to generate identity‑aware captions. Experiments show that ID‑HOINet achieves state‑of‑the‑art performance on the BEHAVE and InterCap datasets.
By Yiming Wang, Yonghao Dang, Huilai Li, Jiawei Tu, Jianqin Yin