Kailun Yang
PulseAugur coverage of Kailun Yang — every cluster mentioning Kailun Yang across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
HierDAMap framework advances universal domain adaptive BEV mapping
Researchers have introduced HierDAMap, a novel framework designed for universal domain adaptive Bird's-Eye View (BEV) mapping. This approach utilizes hierarchical perspective priors to enhance unsupervised domain adapta…
-
New dataset and framework for panoramic robot perception unveiled
Researchers have introduced PanoMMOcc, a new dataset and framework for panoramic multimodal semantic occupancy prediction specifically designed for quadruped robots. The proposed VoxelHound framework addresses challenge…
-
New framework PanoAffordanceNet tackles 360-degree affordance grounding
Researchers have introduced PanoAffordanceNet, a novel framework designed for holistic affordance grounding in 360-degree indoor environments. This new task addresses challenges such as geometric distortions from equire…
-
New benchmark OccTrack360 advances 4D occupancy tracking with fisheye cameras
Researchers have introduced OccTrack360, a new benchmark designed for 4D panoptic occupancy tracking using surround-view fisheye cameras. This benchmark addresses limitations in existing datasets by providing longer seq…
-
New benchmark and method tackle noisy labels in 3D semantic occupancy prediction
Researchers have introduced OccNL, a new benchmark designed to evaluate 3D semantic occupancy prediction models under noisy label conditions. They found that existing 2D label noise learning strategies perform poorly in…
-
New Gaussian-based methods advance 3D scene reconstruction
Researchers have developed new methods for 3D scene reconstruction using Gaussian Opacity Fields and 3D Gaussian Triangulation. Spherical-GOF adapts 3D Gaussian Splatting for omnidirectional images, improving geometric …
-
NOVA model uses LLMs for advanced 3D multi-object tracking in autonomous driving
Researchers have introduced NOVA, a novel autoregressive formulation for 3D Multi-Object Tracking (3D MOT) designed to overcome the limitations of closed-set assumptions in autonomous driving perception. NOVA reformulat…
-
New dataset and model tackle multi-person 3D motion editing
Researchers have introduced InterEdit3D, a new dataset and benchmark for text-guided 3D motion editing involving multiple people. The proposed InterEdit model, a diffusion model, addresses the challenges of limited pair…
-
New FiCoP Framework Enhances Robot Object Pose Estimation
Researchers have developed a new framework called FiCoP to improve open-vocabulary 6D object pose estimation, a capability crucial for robots to manipulate unseen objects using natural language. FiCoP addresses limitati…
-
FoundCAC framework improves lens aberration correction with AI
Researchers have developed FoundCAC, a new framework for blind lens aberration correction that utilizes large-scale pre-training and discrete degradation priors. The system improves data scalability by constructing a di…
-
LFX framework unifies light field data for segmentation and detection
Researchers have introduced LFX, a novel unified framework designed to handle various light field (LF) representations for dense semantic segmentation and salient object detection. This framework utilizes a representati…
-
OmniTrack++ advances panoramic multi-object tracking with trajectory feedback
Researchers have introduced OmniTrack++, an advanced framework for omnidirectional multi-object tracking designed to overcome challenges like panoramic distortion and identity ambiguity. The system utilizes a feedback-d…