Zobrazeno 1 - 10
of 450
pro vyhledávání: '"Zhu, Chenyang"'
Reconstructing from multi-view images is a longstanding problem in 3D vision, where neural radiance fields (NeRFs) have shown great potential and get realistic rendered images of novel views. Currently, most NeRF methods either require accurate camer
Externí odkaz:
http://arxiv.org/abs/2411.02979
Radiance fields including NeRFs and 3D Gaussians demonstrate great potential in high-fidelity rendering and scene reconstruction, while they require a substantial number of posed images as inputs. COLMAP is frequently employed for preprocessing to es
Externí odkaz:
http://arxiv.org/abs/2408.16690
Autor:
Li, Wenhao, Yu, Zhiyuan, She, Qijin, Yu, Zhinan, Lan, Yuqing, Zhu, Chenyang, Hu, Ruizhen, Xu, Kai
The household rearrangement task involves spotting misplaced objects in a scene and accommodate them with proper places. It depends both on common-sense knowledge on the objective side and human user preference on the subjective side. In achieving su
Externí odkaz:
http://arxiv.org/abs/2408.12093
Low-level 3D representations, such as point clouds, meshes, NeRFs, and 3D Gaussians, are commonly used to represent 3D objects or scenes. However, humans usually perceive 3D objects or scenes at a higher level as a composition of parts or structures
Externí odkaz:
http://arxiv.org/abs/2408.10789
Combinatorial Optimization (CO) problems are fundamentally important in numerous real-world applications across diverse industries, characterized by entailing enormous solution space and demanding time-sensitive response. Despite recent advancements
Externí odkaz:
http://arxiv.org/abs/2406.19705
The insertion of objects into a scene and relighting are commonly utilized applications in augmented reality (AR). Previous methods focused on inserting virtual objects using CAD models or real objects from single-view images, resulting in highly lim
Externí odkaz:
http://arxiv.org/abs/2406.14806
This paper focuses on training a robust RGB-D registration model without ground-truth pose supervision. Existing methods usually adopt a pairwise training strategy based on differentiable rendering, which enforces the photometric and the geometric co
Externí odkaz:
http://arxiv.org/abs/2405.00507
This paper introduces MultiBooth, a novel and efficient technique for multi-concept customization in image generation from text. Despite the significant advancements in customized generation methods, particularly with the success of diffusion models,
Externí odkaz:
http://arxiv.org/abs/2404.14239
This work focuses on the dual-arm object rearrangement problem abstracted from a realistic industrial scenario of Cartesian robots. The goal of this problem is to transfer all the objects from sources to targets with the minimum total completion time
Externí odkaz:
http://arxiv.org/abs/2402.13634
Learning-based surface reconstruction based on unsigned distance functions (UDF) has many advantages such as handling open surfaces. We propose SuperUDF, a self-supervised UDF learning which exploits a learned geometry prior for efficient training an
Externí odkaz:
http://arxiv.org/abs/2308.14371