Handoff: pretrain on HelixNet for the Kirioll highway classifier

2026-07-10 16:04 · handoff for a fresh agent on another machine (no vault access) · from Claude Code session "PCD segmentation deep research"

Continuation target

Produce a supervised-pretrained point-cloud semantic segmentation checkpoint on HelixNet (the only fully license-verified, commercially usable dense-label LiDAR dataset we found — CC-BY-4.0) that we will later fine-tune on 2–5 km of our own annotated highway MLS corridor. The checkpoint, its exact config, and its input-feature contract are the deliverables — not a usable highway classifier by itself.

Context (all you need; the full research lives elsewhere)

Dataset: HelixNet

ItemValue
Zenodo record (license source of truth)zenodo.org/records/6519817 — rights: CC-BY-4.0
PaperLoiseau, Aubry, Landrieu, Online Segmentation of LiDAR Sequences: Dataset and Algorithm, ECCV 2022 — arXiv:2206.08194
Official repo (dataloaders, class defs)github.com/romainloiseau/HelixNet
Content~10 billion points, dense automotive LiDAR sequences with per-point semantic labels + sensor rotation/fiber/timestamp attributes
Size / classesCheck on the Zenodo record before downloading — reserve disk accordingly (order of 100s of GB); take the class list from the official repo, do not guess it

Steps

  1. Environment: Linux + CUDA GPU. PTv3 path needs Ampere-or-newer (FlashAttention) + CUDA ≥ 11.6, PyTorch ≥ 1.12; SPT path runs on 16 GB GPUs (official 11 GB configs exist).
    git clone https://github.com/Pointcept/Pointcept        # PTv3 path (MIT)
    # or
    git clone https://github.com/drprojects/superpoint_transformer  # SPT path (MIT)
  2. Download HelixNet from the Zenodo record (use zenodo_get 6519817 or the per-file URLs). Verify checksums.
  3. Inspect the native format with the official HelixNet repo's dataloaders before converting. Note which attributes exist (xyz, intensity, ring/fiber, time, label).
  4. Convert to the trainer's layout. Pointcept's DefaultDataset wants per-sample dirs of .npy arrays named coord, strength (intensity), segment (labels); set feat_keys/in_channels to match. Critical contract: use geometry + intensity only (no RGB) — our highway MLS has no RGB, and the fine-tune must reuse the same input channels or the first layer breaks.
  5. Sanity run: overfit ~10 sweeps until near-zero train loss — proves the data pipeline before burning GPU-days.
  6. Train on a train/val split by sequence (not random per-sweep!). Starting hyperparameters (extrapolated from the frameworks' outdoor/DALES configs, not gospel): grid size 0.05 m, CE + Lovász loss, AdamW, OneCycle; batch to GPU memory. Log per-class IoU, not just mean.
  7. Record intensity normalization (e.g. clip + scale constants) — it is part of the checkpoint contract for later fine-tuning.

Expected results — calibrate before you start

WhatExpectationBasis
Val mIoU on HelixNet itself Should land in the same family as published outdoor per-sweep benchmarks (60s–70s mIoU); treat the paper's own baselines as the reference once you read them No verified PTv3/SPT-on-HelixNet number exists — do not promise a target; measure and report
Value for the low-label fine-tune (the actual point) Large gains in the small-label regime: published pretraining results show ~+14 mIoU at 1% labels (56.3 vs 42.2, CLIP2Scene on nuScenes) and 45.3-vs-26.0 (Sonata, indoor); at 100% labels the gain shrinks to ~1–2 points arXiv:2301.04926, arXiv:2503.16429
Convergence speed downstream Fine-tuning from this checkpoint should converge in a fraction of the epochs of from-scratch training Standard transfer-learning behavior reported across the pretraining literature above
Which classes transfer Strong: road/ground, vegetation, poles, barriers, vehicles (geometry-dominant, present in HelixNet). Weak/none: Kirioll-specific rare classes (noise barriers, drains, shafts, call boxes) — those come from our own labels later, pretraining only provides better features Class overlap analysis from the research phase
Honest unknown HelixNet is per-sweep automotive LiDAR; our target is dense accumulated survey MLS (~10–50× denser). Domain-gap papers report the shift is real but bridgeable (adaptation techniques recover up to +20 mIoU). The transfer gain could shrink — the first Kirioll fine-tune is the measurement, and grid-sampling both domains to the same voxel size is the first mitigation arXiv:2003.01174, arXiv:2407.06043

What this pretraining does NOT produce: a usable highway classifier. Out of the box the checkpoint segments HelixNet's classes on automotive-style sweeps. Its value is as a warm start that makes our 2–5 km of hand-labeled highway data go 5–20× further.

Deliverables & verification

Risks / open questions