Graphics

Davis Rempe, Mathis Petrovich, Ye Yuan et al.

This addresses the need for scalable, high-quality human motion data for applications in robotics, simulation, and entertainment, representing a significant advancement over previous limited datasets.

Tianshi Cao, Jiawei Ren, Yuxuan Zhang et al.

For autonomous vehicle developers, it addresses the bottleneck of generating simulation-ready 3D assets from sparse in-the-wild data, enabling more realistic closed-loop simulation.

Junxuan Li, Rawal Khirodkar, Chengan He et al.

This work addresses the problem of creating realistic and generalizable 3D avatars for applications in virtual reality, gaming, and digital humans, representing a novel approach rather than an incremental improvement.

Weijie Wang, Qihang Cao, Sensen Gao et al.

For researchers in computer vision and graphics, this survey provides a structured overview of feed-forward 3D reconstruction methods, highlighting common design patterns and open challenges.

Kaixuan Wang, Tianxing Chen, Jiawei Liu et al.

This addresses the problem of limited simulation data for robotic manipulation researchers, though it is incremental as it builds on existing simulation-based learning paradigms.

Xilong Zhou, Bao-Huy Nguyen, Zheng Zeng et al.

For video editing practitioners, AlbedoEdit provides a single model capable of multiple fine-grained editing tasks with superior quality, addressing the lack of unified solutions.

Ryan Po, David Junhao Zhang, Amir Hertz et al.

This work addresses the lack of user control and shared inference in video world models, enabling editable and multiplayer experiences for game developers and interactive simulation users.

Guikun Chen, Wenguan Wang

It addresses the need for a comprehensive overview of this emerging method for researchers in computer graphics and vision, but it is incremental as it surveys existing developments rather than introducing new findings.

Tianhe Wu, Kun Yan, Zikai Zhou et al.

For researchers working on accelerating text-to-image and image editing models, this work provides empirical insights into training recipe design for few-step distillation, though it is incremental in nature.

Dilin Wang, Xiaoyu Xiang, Kihyuk Sohn et al.

It addresses the need for fast, deployable 3D asset generation for interactive workflows, particularly in gaming and AR/VR, where user experience and real-time performance are critical.

Jiahui Huang, Jiawei Ren, Michal Tyszkiewicz et al. · nvidia

This work addresses the need for fast, per-scene tuning-free neural reconstruction for autonomous driving simulation, enabling efficient closed-loop policy evaluation.

Jaskirat Singh, Boyang Zheng, Zongze Wu et al.

This work provides a practical, training-efficient improvement for generative modeling with diffusion transformers, relevant to researchers in image and video generation.

Jon Hasselgren, Zheng Zeng, Milos Hasan et al.

This addresses the problem of efficient material generation for 3D content creators, though it appears incremental as it builds on existing diffusion and VAE techniques.

Shuzhao Xie, Junchen Ge, Weixiang Zhang et al.

For practitioners deploying 3DGS in storage-constrained environments, this method provides a size-aware codec that accurately meets target budgets without retraining.

Kuangshi Ai, Haichao Miao, Kaiyuan Tang et al.

This addresses the need for reproducible evaluation in the scientific visualization community, though it is incremental as it builds on existing agentic systems.

Aviad Dahan, Moran Yanuka, Noa Kraicer et al. · apple-ml

It addresses the challenge of synchronizing personalized audio with video for content creators, offering a novel integrated approach rather than incremental improvements.

Jen-Hao Rick Chang, Xiaoming Zhao, Dorian Chan et al.

This work addresses the challenge of capturing realistic view-dependent effects like specular highlights in 3D reconstruction and generation for computer vision and graphics applications, representing a novel method for a known bottleneck.

Timo Teufel, Xilong Zhou, Umar Iqbal et al.

This work addresses the lack of expressive facial animation in relightable full-body avatars, which is crucial for telepresence, gaming, and virtual production applications.

Etai Sella, Hao Phung, Nitay Amiel et al.

It addresses the challenge of fine-grained 3D editing for users needing precise structural modifications without altering overall object identity, offering a novel approach that avoids costly training.

Lin Song, Wenbo Li, Guoqing Ma et al.

This work advances unified visual models for researchers aiming to integrate understanding and generation with spatial reasoning, though the approach is incremental as it builds on existing MLLM and diffusion architectures.