← Back to the feed
AI & softwareResearch findings

TeFF · A Tracking-Enhanced Method for Few-Shot 3D LiDAR Semantic Segmentation

Original published 2024.08.28 18:18 KSTAdded to Blackdeck 2026.11.13 03:28 KST

What has changed?

A tracking model generates pseudo-ground-truths from LiDAR frame sequences to enhance few-shot learning.

Key points

  1. Proposes a tracking-enhanced approach to address few-shot learning in 3D LiDAR semantic segmentation.
  2. Generates pseudo-ground-truths from LiDAR frame sequences using a tracking model to expand training data.
  3. Applies LoRA to preserve base-class performance while improving adaptation to novel classes.

Connect it to your work

Temporal continuity of LiDAR data enables effective pseudo-label generation for novel object learning

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivTeFF: Tracking-enhanced Forgetting-free Few-shot 3D LiDAR Semantic Segmentation