Paper Detail

Towards Real-Time Full-Waveform LiDAR Transformers via Intensity-Guided Token Reduction and Physics-Aware Augmentation

Kotaro Oishi, Kazuma Ikeda, Ryosei Hara, Ryo Yoshida, Mariko Isogawa, Kentaro Yoshioka

huggingface Score 7.4

Published 2026-11-05 · First seen 2026-11-07

General AI

Abstract

LiDAR is a critical sensor for autonomous driving and robotics, yet it remains vulnerable to adverse conditions such as fog, rain, transparent objects, and low-reflectance surfaces. Full-waveform LiDAR (FWL) addresses these limitations by capturing the complete return waveform as a histogram, preserving rich reflection characteristics including multi-path echoes and weak signals. While recent Transformer-based approaches have demonstrated promise on FWL data, two fundamental bottlenecks remain: high computational cost from redundant tokens in low-intensity background regions, and limited generalization due to scarce annotated training data. We address both bottlenecks by exploiting the physical structure inherent to FWL data. First, we propose FWL-ToPM, a token pruning and merging mechanism that leverages intensity distributions to selectively reduce tokens while retaining critical waveform content. Second, we introduce FWLAug, a data augmentation framework designed as a regularization method that preserves waveform structure, retaining the Time-of-Flight waveform of each ray and the frustum geometry, to improve generalization to unseen scenes. Evaluated on the Ghost-FWL benchmark, FWL-ToPM achieves over 8 times speedup at real-time throughput alongside a nearly 9-point F1 gain over the prior state of the art, establishing the first real-time FWL Transformer that simultaneously improves accuracy. FWLAug further improves performance on both the standard benchmark and an out-of-distribution test set of unseen scenes. Together, these methods push FWL-based perception toward practical deployment. The code and the OOD dataset are publicly available and can be accessed via the project page: https://keio-csg.github.io/FWL-ToPM

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@misc{oishi2026real,
  title = {Towards Real-Time Full-Waveform LiDAR Transformers via Intensity-Guided Token Reduction and Physics-Aware Augmentation},
  author = {Kotaro Oishi and Kazuma Ikeda and Ryosei Hara and Ryo Yoshida and Mariko Isogawa and Kentaro Yoshioka},
  year = {2026},
  abstract = {LiDAR is a critical sensor for autonomous driving and robotics, yet it remains vulnerable to adverse conditions such as fog, rain, transparent objects, and low-reflectance surfaces. Full-waveform LiDAR (FWL) addresses these limitations by capturing the complete return waveform as a histogram, preserving rich reflection characteristics including multi-path echoes and weak signals. While recent Transformer-based approaches have demonstrated promise on FWL data, two fundamental bottlenecks remain: },
  url = {https://huggingface.co/papers/2611.07613},
  keywords = {Full-waveform LiDAR (FWL), Transformer, token pruning and merging, FWL-ToPM, FWLAug, waveform-preserving data augmentation, regularization, Time-of-Flight waveform, frustum geometry, out-of-distribution generalization, Ghost-FWL benchmark, real-time FWL Transformer, intensity distributions, multi-path echoes, weak signals, code available, huggingface daily},
  eprint = {2611.07613},
  archiveprefix = {arXiv},
}

Metadata

{}