TFusionOcc: A T-Primitive Based Object-Centric Multi-Sensor Fusion Framework for 3D Occupancy Prediction
TFusionOcc: T-Primitive Based Object-Centric Multi-Sensor Fusion Framework for 3D Occupancy Prediction
This research presents TFusionOcc, a novel framework for 3D semantic occupancy prediction that utilizes a family of Student's t-distribution-based T-primitives (including plain T-primitive, T-Superquadric, and deformable T-Superquadric) to improve representation efficiency for complex structures. The authors develop a unified probabilistic formulation using a T-mixture model to jointly model occupancy and semantics, alongside a tightly coupled multi-stage fusion architecture integrating camera and LiDAR data. Experiments on the nuScenes dataset show competitive performance against state-of-the-art methods, with robustness tested on nuScenes-C under corruption scenarios and cross-dataset applicability demonstrated on SemanticKITTI. The code is available at https://github.com/DanielMing123/TFusionOcc.
The paper introduces TFusionOcc, a framework using Student's t-distribution-based T-primitives for 3D semantic occupancy prediction, evaluated on nuScenes, nuScenes-C, and SemanticKITTI benchmarks.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- code: Reported pending Open source S5
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ly. Extensive experiments on nuScenes demonstrate competitive performance against recent state-of-the-art methods and strong performance among object-centric multi-sensor approaches, while additional evaluations on nuScenes-C show robust performance across most corruption scenarios. Further evaluation on the separate SemanticKITTI benchmark demonstrates cross-dataset applicability. The code will be available at: http
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org