Consistency-aware Self-Training for Iterative-based Stereo Matching
CST-Stereo achieves impressive results in various scenarios including in-domain, domain adaptive and domain generalization.
jingyi-zhou
•
•
1 min read
GeoX: Geometric Problem Solving Through Unified Formalized Vision-Language Pre-training
A multi-modal large model focusing on geometric understanding and reasoning tasks with formalized visual-language pre-training.
renqiu-xia
All-in-One: Transferring Vision Foundation Models into Stereo Matching
AIOStereo flexibly selects and transfers knowledge from multiple heterogeneous VFMs to a single stereo matching model. Rank 1st on Middlebury Stereo Evaluation.
jingyi-zhou
Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy
AdaptiveDiffusion adaptively reduces noise prediction steps during denoising guided by third-order latent difference.
hancheng-ye
3DET-Mamba: Causal Sequence Modelling for End-to-End 3D Object Detection
Exploit the potential of Mamba architecture on 3D scene-level perception for the first time.
mingsheng-li
Reg-TTA3D: Better Regression Makes Better Test-time Adaptive 3D Object Detection
A pseudo-label-based test-time adaptive 3D object detection method exploring a new task of test-time domain adaptive 3D detection.
jiakang-yuan
ReSimAD: Zero-Shot 3D Domain Transfer for Autonomous Driving with Source Reconstruction and Target Simulation
A Reconstruction-Simulation-Perception scheme for alleviating domain shifts in autonomous driving.
bo-zhang
AD-PT: Autonomous Driving Pre-Training with Large-scale Point Cloud Dataset
Build a large-scale pre-training point-cloud dataset with diverse data distribution, and learn generalizable representations.
jiakang-yuan
Uni3D: A Unified Baseline for Multi-dataset 3D Object Detection
A unified baseline to tackle multi-dataset 3D object detection from data-level and semantic-level.
bo-zhang
Bi3D: Bi-domain Active Learning for Cross-domain 3D Object Detection
A Bi-domain active learning approach which selects samples from both source and target domain to solve cross-domain 3D object detection.
jiakang-yuan