Metraipolis › archive › dc_a98934be592b
CRISP: Fixing Flying Pixels in Latent LiDAR Generation via Diffusion Decoding — arXiv:2610.11376
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.11376
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T05:57:39.506Z
- SHA-256 of the archived text
4eb9282f2c7015b52f36cebeeb383ba1dd475793606183aaa5139611a18fd5b0- Stable URL of this page
https://metraipolis.com/documento/dc_a98934be592b
How to cite
CRISP: Fixing Flying Pixels in Latent LiDAR Generation via Diffusion Decoding — arXiv:2610.11376 — arXiv (en), https://arxiv.org/abs/2610.11376 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a98934be592b (Metraipolis archive dc_a98934be592b, sha256 4eb9282f2c7015b52f36cebeeb383ba1dd475793606183aaa5139611a18fd5b0)
BibTeX
@misc{dca98934be592b,
title = {CRISP: Fixing Flying Pixels in Latent LiDAR Generation via Diffusion Decoding — arXiv:2610.11376},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.11376},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a98934be592b; sha256 4eb9282f2c7015b52f36cebeeb383ba1dd475793606183aaa5139611a18fd5b0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.11376. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Latent LiDAR pipelines suffer from flying pixels: convolutional VAEs blur sharp radial depth discontinuities, yielding edge depths that back-project to points floating between surfaces. We identify this as a major, directly correctable decoder bottleneck and introduce CRISP: a pixel-space diffusion decoder with a backbone-agnostic latent adapter, DiT-based denoiser, and support mask predictor. CRISP replaces video-VAE and LiDAR-native decoders alike while keeping the encoder and latent generator fixed. Across KITTI-360, SemanticKITTI, and nuScenes, replacing only the decoder reduces FSVD/FPVD by 50.5% on average across frozen backbones; for generic video VAEs, the reductions reach 71%/74%. On the LiDAR-native LiDM backbone, FRID drops by 71%, with the largest gains at depth discontinuities. In a pretrained LiDM world model, the same zero-shot replacement improves FSVD by 15.5%, narrowing the sim-to-real gap. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.11376 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-09T05:47:01.614Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).