7 papers · 1 filter
Beyond Global Latents: Chunk-Based Sparse Grid VAE for Scalable 3D Modeling
Kaiyi Zhang, Zhihao Liang, Haolin Liu +8
Sparse voxel grids preserve the spatial structure needed for detailed 3D reconstruction, but their memory still grows rapidly with resolution as active surface cells increase. We i…
MSVS-VAE: Multi-Scale Anchored VecSet for High-Fidelity 3D Reconstruction
Dehao Hao, Kaiyi Zhang, Tanghui Jia +10
High-fidelity 3D generative modeling increasingly relies on the latent diffusion paradigm, where the reconstruction quality of the underlying 3D VAE becomes a primary bottleneck. E…
Map2Thought: Explicit 3D Spatial Reasoning via Metric Cognitive Maps
Xiangjun Gao, Zhensong Zhang, Dave Zhenyu Chen +4
We propose Map2Thought, a framework that enables explicit and interpretable spatial reasoning for 3D VLMs. The framework is grounded in two key components: Metric Cognitive Map (Me…
Complex-Valued 2D Gaussian Representation for Computer-Generated Holography
Yicheng Zhan, Xiangjun Gao, Long Quan +1
Complex-valued Gaussian primitives have recently been explored for representing holographic radiance fields in 3D novel view synthesis. In this work, we extend this line of researc…
DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos
Wenbo Hu, Xiangjun Gao, Xiaoyu Li +5
Estimating video depth in open-world scenarios is challenging due to the diversity of videos in appearance, content motion, camera movement, and length. We present DepthCrafter, an…
ConTex-Human: Free-View Rendering of Human from a Single Image with Texture-Consistent Synthesis
Xiangjun Gao, Xiaoyu Li, Chaopeng Zhang +4
In this work, we propose a method to address the challenge of rendering a 3D human from a single image in a free-view manner. Some existing approaches could achieve this by using g…