Neural Scene Representation and Neural Rendering

Seminar – Fall Semester 2025

Instructor: Lingjie Liu

TAs: Chuhao Chen, Abhimanyu Suthar

Organization | Topics | Format | Resources


NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis ECCV, 2020	NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction NeurIPS, 2021

Topics

Paper Presenting Schedule	Presentation
Introduction slides (by Lingjie)	Sep 2
Introduction 2 slides (by Lingjie)	Sep 4
Taming 3DGS: High-Quality Radiance Fields with Limited Resources Mallick, Goel et al. SIGGRAPH Asia 2024 + DashGaussian: Optimizing 3D Gaussian Splatting in 200 Seconds Chen et al. CVPR 2025 (Highlight) slides MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images Chen et al. ECCV 2024 (Oral) + Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Image Szymanowicz, Insafutdinov et al. 3DV 2025 slides	Sep 9 Presenter 1: Ningze Zhong Presenter 2: Michael Jacob
3D Student Splatting and Scooping Zhu et al. CVPR 2025 (Oral, Best Paper Honourable Mention) + Volumetrically Consistent 3D Gaussian Rasterization Talegaonkar et al. CVPR 2025 (Highlight) slides Quark: Real-time, High-resolution, and General Neural View Synthesis Flynn, Broxton, Murmann et al. SIGGRAPH Asia 2024 (Best Paper Award) slides	Sep 11 Presenter 1: Dixuan Lin Presenter 2: Quynh Anh Huynh
DUSt3R: Geometric 3D Vision Made Easy Wang et al. CVPR 2024 + SPANN3R: 3D Gaussian Splatting for Large-Scale Neural Reconstruction in the Wild Wang et al. 3DV 2025 + Fast3R: Fast and High-Quality 3D Gaussian Splatting for Radiance Fields Yang et al. CVPR 2025 slides VGGSfM: Accurate, Fast, and Scalable Structure-from-Motion Wang et al. CVPR 2024 (Highlight) slides	Sep 16 Presenter 1: Shubhan Pawar Presenter 2: Samuel Chua
VGGT: Visual Geometry Grounded Transformer Wang et al. CVPR 2025 (Best Paper Award) slides π³: Scalable Permutation-Equivariant Visual Geometry Learning Wang, Zhou et al. arXiv 2025 slides	Sep 18 Presenter 1: Siyu Li Presenter 2: Nehir Sunar
LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias Jin et al. ICLR 2025 (Oral) slides RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination Zeng et al. SIGGRAPH 2025 slides	Sep 23 Presenter 1: Aileen Liao Presenter 2: Zhaowen Gu
MeshFormer: High-Quality Mesh Generation with 3D-Guided Reconstruction Model Liu, Zeng et al. NeurIPS 2024 (Oral) slides TRELLIS: Structured 3D Latents for Scalable and Versatile 3D Generation Xiang et al. CVPR 2025 (Highlight) slides	Sep 25 Presenter 1: Shubhan Pawar Presenter 2: Daiyuan Li
SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency Xie et al. ICLR 2025 + SV4D 2.0: Enhancing Spatio-Temporal Consistency in Multi-View Video Diffusion for High-Quality 4D Generation Yao et al. ICCV 2025 + CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models Wu et al. CVPR 2025 slides MoSca: Dynamic Gaussian Fusion from Casual Videos via 4D Motion Scaffolds Lei et al. CVPR 2025 (Highlight) slides	Sep 30 Presenter 1: Yuntian Ke Presenter 2: Neel Shejwalkar
Shape of Motion: 4D Reconstruction from a Single Video Wang, Ye, Gao, Zeng et al. ICCV 2025 (Highlight) slides DIMO: Diverse 3D Motion Generation for Arbitrary Objects Mou et al. ICCV 2025 (Highlight) slides EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting Lee et al. CVPR 2025 slides	Oct 2 Presenter 1: Sagnik Anupam Presenter 2: Tianang Leng Presenter 3: Siyu Li
Guest Talk: Physics-Grounded World Models: Generation, Interaction, and Evaluation slides	Oct 7 Presenter: Hong-Xing "Koven" Yu (remote)
No Class	Oct 9
Guest Talk: Building Scalable, Human-Centric Physical AI Systems slides	Oct 14 Presenter: Wayne Wu (in-person)
Reconstruction and Simulation of Elastic Objects with Spring-Mass 3D Gaussians Zhong et al. ECCV 2024 + PhysTwin: Physics-Informed Reconstruction and Simulation of Deformable Objects from Videos Jiang et al. ICCV 2025 slides Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation Chen et al. CVPR 2025 slides PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation Wang, Chen et al. NeurIPS 2025 slides	Oct 16 Presenter 1: Quynh Anh Huynh Presenter 2: Daiyuan Li Presenter 3: Herbie Wright
No Class	Oct 21 - Oct 23
Guest Talk: Sparse View Synthesis	Oct 28 Presenter: Ravi Ramamoorthi
Dr. Splat: Directly Referring 3D Gaussian Splatting via Direct Language Embedding Registration Kim, Kim et al. CVPR 2025 (Highlight) slides A Hierarchical 3D Gaussian Representation for Real-Time Rendering of Very Large Datasets Kerbl, Meuleman et al. SIGGRAPH 2024 + On-the-fly Reconstruction for Large-Scale Novel View Synthesis from Unposed Images Meuleman et al. SIGGRAPH 2025 slides PhysMotion: Physics-Grounded Dynamics From a Single Image Tan, Jiang, Li et al. arXiv 2024 + WonderPlay: Dynamic 3D Scene Generation from a Single Image and Actions Li, Yu et al. ICCV 2025 (Highlight) slides	Oct 30 Presenter 1: Zhaowen Gu Presenter 2: Kevin Liu Presenter 3: Ningze Zhong
MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors Murai, Dexheimer et al. CVPR 2025 (Highlight) slides GWM: Towards Scalable Gaussian World Models for Robotic Manipulation Lu , Jia, Li et al. ICCV 2025 slides D(R,O) Grasp: A Unified Representation of Robot and Object Interaction for Cross-Embodiment Dexterous Grasping Wei, Xu et al. ICRA 2025 (Best Paper Award) slides	Nov 4 Presenter 1: Neel Shejwalkar Presenter 2: Herbie Wright Presenter 3: Samuel Chua
Splat-Nav: Safe Real-Time Robot Navigation in Gaussian Splatting Maps Chen, Shorinwa et al. TRO 2025 Pre-training Auto-regressive Robotic Models with 4D Representations Niu, Sharama et al. ICML 2025 Gaussian Splashing: Unified Particles for Versatile Motion Synthesis and Rendering Feng, Feng et al. CVPR 2025	Nov 6 Presenter 1: Sagnik Anupam Presenter 2: Aileen Liao Presenter 3: Hanqi Su
Objects as volumes: A stochastic geometry view of opaque solids Miller et al. CVPR 2024 Geometry Field Splatting with Gaussian Surfels Jiang et al. CVPR 2025 SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering Guédonet et al. CVPR 2024	Nov 11 Presenter 1: Hanqi Su Presenter 2: Nehir Sunar Presenter 3: Luyang Hu
CAST: Component-Aligned 3D Scene Reconstruction from an RGB Image Yao, Zhang et al. SIGGRAPH 2025 (Best Paper Award) HoloPart: Generative 3D Part Amodal Segmentation Yang et al. arXiv 2025 + PartField: Learning 3D Feature Fields for Part Segmentation and Beyond Liu, Uy et al. ICCV 2025 ARM: Appearance Reconstruction Model for Relightable 3D Generation Feng, Yu, Bi et al. CVPR 2025 (Highlight) RelitLRM: Generative Relightable Radiance for Large Reconstruction Models Zhang et al. ICLR 2025 (Spotlight)	Nov 13 Presenter 1: Yuntian Ke Presenter 2: Michael Jacob Presenter 3: Kevin Liu
GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control Ren, Shen et al. CVPR 2025 (Highlight) Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models Wu, Zhang et al. CVPR 2025 (Best Paper Honorable Mention) Voyaging into Perpetual Dynamic Scenes from a Single View Tian et al. ICCV 2025	Nov 18 Presenter 1: Tianang Leng Presenter 2: Dixuan Lin Presenter 3: Luyang Hu
Favorite Papers & Brain Storming	Nov 20
Practice Lecture & Summary	Nov 25
No Class	Nov 27
No Class (Tentative)	Dec 2
No Class (Tentative)	Dec 4

The following table lists all the topics and papers that will be discussed in the seminar. Once every participant has submitted their choice of topics and papers, the previous list will be updated to show the presenter of each topic. Send us an email if you cannot access a paper for some reason.

Click on each topic to show the papers to be discussed or show all papers hide all papers.

Topic and Papers
Fast Training Taming 3DGS: High-Quality Radiance Fields with Limited Resources Mallick, Goel et al. SIGGRAPH Asia 2024 + DashGaussian: Optimizing 3D Gaussian Splatting in 200 Seconds Chen et al. CVPR 2025 (Highlight)
Generalisable 3D Reconstruction MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images Chen et al. ECCV 2024 (Oral) + Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Image Szymanowicz, Insafutdinov et al. 3DV 2025
High-quality Rendering 3D Student Splatting and Scooping Zhu et al. CVPR 2025 (Oral, Best Paper Honorable Mention) + Volumetrically Consistent 3D Gaussian Rasterization Talegaonkar et al. CVPR 2025 (Highlight) Quark: Real-time, High-resolution, and General Neural View Synthesis Flynn, Broxton, Murmann et al. SIGGRAPH Asia 2024 (Best Paper Award)
Fast Visual Geometry DUSt3R: Geometric 3D Vision Made Easy Wang et al. CVPR 2024 + Spann3R: 3D Reconstruction with Spatial Memory Wang et al. 3DV 2025 + Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass Yang et al. CVPR 2025 VGGSfM: Visual Geometry Grounded Deep Structure From Motion Wang et al. CVPR 2024 (Highlight) VGGT: Visual Geometry Grounded Transformer Wang et al. CVPR 2025 (Best Paper Award) Pi^3: Scalable Permutation-Equivariant Visual Geometry Learning Wang, Zhou et al. arXiv 2025
Transformer-based Neural Rendering LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias Jin et al. ICLR 2025 (Oral) RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination Zeng et al. SIGGRAPH 2025
3D Generation MeshFormer: High-Quality Mesh Generation with 3D-Guided Reconstruction Model Liu, Zeng et al. NeurIPS 2024 (Oral) TRELLIS: Structured 3D Latents for Scalable and Versatile 3D Generation Xiang et al. CVPR 2025 (Highlight)
4D Reconstruction & Generation SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency Xie et al. ICLR 2025 + SV4D 2.0: Enhancing Spatio-Temporal Consistency in Multi-View Video Diffusion for High-Quality 4D Generation Yao et al. ICCV 2025 + CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models Wu et al. CPPR 2025 MoSca: Dynamic Gaussian Fusion from Casual Videos via 4D Motion Scaffolds Lei et al. CVPR 2025 (Highlight) Shape of Motion: 4D Reconstruction from a Single Video Wang, Ye, Gao, Zeng et al. ICCV 2025 (Highlight) DIMO: Diverse 3D Motion Generation for Arbitrary Objects Mou et al. ICCV 2025 (Highlight)
Physics-grounded Reconstruction & Generation Reconstruction and Simulation of Elastic Objects with Spring-Mass 3D Gaussians Zhong et al. ECCV 2024 + PhysTwin: Physics-Informed Reconstruction and Simulation of Deformable Objects from Videos Jiang et al. ICCV 2025 Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation Chen et al. CVPR 2025 Gaussian Splashing: Unified Particles for Versatile Motion Synthesis and Rendering Feng, Feng et al. CVPR 2025 PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation Wang, Chen et al. NeurIPS 2025 PhysMotion: Physics-Grounded Dynamics From a Single Image Tan, Jiang, Li et al. arXiv 2024 + WonderPlay: Dynamic 3D Scene Generation from a Single Image and Actions Li, Yu et al. ICCV 2025 (Highlight)
Multi-modality Dr. Splat: Directly Referring 3D Gaussian Splatting via Direct Language Embedding Registration Kim, Kim et al. CVPR 2025 (Highlight)
Large Scene Reconstruction A Hierarchical 3D Gaussian Representation for Real-Time Rendering of Very Large Datasets Kerbl, Meuleman et al. SIGGRAPH 2024 + On-the-fly Reconstruction for Large-Scale Novel View Synthesis from Unposed Images Meuleman et al. SIGGRAPH 2025
Robotics MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors. Murai, Dexheimer et al. CVPR 2025 (Highlight & Best Demo Honorable Mention) GWM: Towards Scalable Gaussian World Models for Robotic Manipulation Lu, Jia, Li et al. ICCV 2025 Splat-Nav: Safe Real-Time Robot Navigation in Gaussian Splatting Maps Chen, Shorinwa et al. TRO 2025 Pre-training Auto-regressive Robotic Models with 4D Representations Niu, Sharma et al. ICML 2025 D(R,O) Grasp: A Unified Representation of Robot and Object Interaction for Cross-Embodiment Dexterous Grasping Wei, Xu et al. ICRA 2025 (Best Paper Award)
Surface Reconstruction Objects as Volumes: A Stochastic Geometry View of Opaque Solids Miller et al. CVPR 2024 Geometry Field Splatting with Gaussian Surfels Jiang et al. CVPR 2025 SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering Guédonet et al. CVPR 2024
Scene Components Reconstruction CAST: Component-Aligned 3D Scene Reconstruction from an RGB Image Yao, Zhang et al. SIGGRAPH 2025 (Best Paper Award)
Part Segmentation HoloPart: Generative 3D Part Amodal Segmentation Yang et al. arXiv 2025 + PartField: Learning 3D Feature Fields for Part Segmentation and Beyond Liu, Uy et al. ICCV 2025
Video Generation GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control Ren, Shen et al. CVPR 2025 (Highlight) Voyaging into Perpetual Dynamic Scenes from a Single View Tian et al. ICCV 2025
Diffusion-based 3D Reconstruction DIFIX3D+: Improving 3D Reconstructions with Single-Step Diffusion Models Wu, Zhang et al. CVPR 2025 (Best Paper Honorable Mention)
Appearance Modeling ARM: Appearance Reconstruction Model for Relightable 3D Generation Feng, Yu, Bi et al. CVPR 2025 (Highlight) RelitLRM: Generative Relightable Radiance for Large Reconstruction Models Zhang et al. ICLR 2025 (Spotlight)
Editing EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting Dong In Lee, Hyeongcheol Park, Jiyoung Seo, Eunbyung Park, Hyunje Park, Ha Dam Baek, Sangheon Shin, Sangmin Kim, Sangpil Kim CVPR 2025