11 papers
IRIS: Implicit Rendering Matters for Pose-Free Novel View Synthesis
Wenyu Li, Sidun Liu, Peng Qiao +2
Novel view synthesis from unposed multi-view images remains challenging, as the model must jointly learn scene representations and camera parameters without pose supervision. Exist…
LocusGS: Spatially Grounded Tokens for Feed-Forward 3D Gaussian Splatting
Wenyu Li, Sidun Liu, Tongrui Hu +2
Recent query-based feed-forward 3DGS methods represent a scene using learnable queries, each aggregating multi-view evidence and decoding a group of Gaussians. Ideally, different q…
Muskie: Multi-view Masked Image Modeling for 3D Vision Pre-training
Wenyu Li, Sidun Liu, Peng Qiao +2
We present Muskie, a native multi-view vision backbone designed for 3D vision tasks. Unlike existing models, which are frame-wise and exhibit limited multi-view consistency, Muskie…
Transolver is a Linear Transformer: Revisiting Physics-Attention through the Lens of Linear Attention
Wenjie Hu, Sidun Liu, Peng Qiao +2
Recent advances in Transformer-based Neural Operators have enabled significant progress in data-driven solvers for Partial Differential Equations (PDEs). Most current research has…
Mono3R: Exploiting Monocular Cues for Geometric 3D Reconstruction
Wenyu Li, Sidun Liu, Peng Qiao +1
Recent advances in data-driven geometric multi-view 3D reconstruction foundation models (e.g., DUSt3R) have shown remarkable performance across various 3D vision tasks, facilitated…
Regist3R: Incremental Registration with Stereo Foundation Model
Sidun Liu, Wenyu Li, Peng Qiao +1
Multi-view 3D reconstruction has remained an essential yet challenging problem in the field of computer vision. While DUSt3R and its successors have achieved breakthroughs in 3D re…
Highly Parallelized Reinforcement Learning Training with Relaxed Assignment Dependencies
Zhouyu He, Peng Qiao, Rongchun Li +2
As the demands for superior agents grow, the training complexity of Deep Reinforcement Learning (DRL) becomes higher. Thus, accelerating training of DRL has become a major research…
VoxNeuS: Enhancing Voxel-Based Neural Surface Reconstruction via Gradient Interpolation
Sidun Liu, Peng Qiao, Zongxin Ye +2
Neural Surface Reconstruction learns a Signed Distance Field~(SDF) to reconstruct the 3D model from multi-view images. Previous works adopt voxel-based explicit representation to i…
DistGrid: Scalable Scene Reconstruction with Distributed Multi-resolution Hash Grid
Sidun Liu, Peng Qiao, Zongxin Ye +2
Neural Radiance Field~(NeRF) achieves extremely high quality in object-scaled and indoor scene reconstruction. However, there exist some challenges when reconstructing large-scale…
Deep RAW Image Super-Resolution. A NTIRE 2024 Challenge Survey
Marcos V. Conde, Florin-Alexandru Vasluianu, Radu Timofte +32
This paper reviews the NTIRE 2024 RAW Image Super-Resolution Challenge, highlighting the proposed solutions and results. New methods for RAW Super-Resolution could be essential in…
AbsGS: Recovering Fine Details for 3D Gaussian Splatting
Zongxin Ye, Wenyu Li, Sidun Liu +2
3D Gaussian Splatting (3D-GS) technique couples 3D Gaussian primitives with differentiable rasterization to achieve high-quality novel view synthesis results while providing advanc…