Research map / Detection and segmentation / 3D detection and driving
FIND: Frequency Invariance Disentanglement for Test-Time Adaptation in LiDAR 3D Detection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes FIND, a test-time adaptation method for LiDAR 3D object detection that disentangles frequency-invariant semantic content from domain style to handle domain shift.
Problem
LiDAR-based 3D object detection degrades under domain shifts from different sensor configurations and environments; existing test-time adaptation methods use spatial-domain self-supervision that confounds semantic content with domain style on sparse, non-Euclidean point clouds.
Approach
Shifts the test-time adaptation signal to the frequency domain, separating domain-invariant semantic content from domain-specific style so adaptation does not entangle the two (based only on the title and abstract).
Closest papers by meaning
- MOS: Model Synergy for Test-Time Adaptation on LiDAR-Based 3D Object Detection (ICLR 2025 · similarity 0.89)
- D^3CTTA: Domain-Dependent Decorrelation for Continual Test-Time Adaption of 3D LiDAR Segmentation (CVPR 2025 · similarity 0.88)
- An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains (CVPR 2024 · similarity 0.88)
- CMDA: Cross-Modal and Domain Adversarial Adaptation for LiDAR-Based 3D Object Detection (AAAI 2024 · similarity 0.88)
- DiffuBox: Refining 3D Object Detection with Point Diffusion (NeurIPS 2024 · similarity 0.87)
- LiDAR-PTQ: Post-Training Quantization for Point Cloud 3D Object Detection (ICLR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.