Weihao Gao · 2026 · arXiv (Cornell University)
DOI: 10.48550/arxiv.2610.31228 · cited by 0 works (OpenAlex) · open-access PDF
La segmentation du polyp de dépassement en temps réel dans la colonoscopie assistée par ordinateur est entravée par la difficulté de déployer des modèles de haute précision sous des budgets de latence stricts. Le dépassement du cadre, qui effectue l'inférence de chaque cadre K et réutilise ou interpole le reste, améliore le débit mais introduit le flou de mouvement et la désalignement de la lésion qui dégrade la cohérence temporelle. Pour combler cette lacune, nous proposons le Compensateur du cadre léger (LFC), un module Plug-and-Play de 77K-paramètre généré par un modèle semé différemment, mais sans aucune annotation d'expert. Les expériences sur trois tableaux de référence de
Machine translation (Meta M2M100); the English original follows.
Real-time polyp segmentation in computer-assisted colonoscopy is hampered by the difficulty of deploying high-accuracy models under strict latency budgets. Frame skipping, which performs inference every K frames and reuses or interpolates the rest, improves throughput but introduces motion blur and lesion misalignment that degrade temporal consistency. To close this gap, we propose the Lightweight Frame Compensator (LFC), a 77K-parameter plug-and-play module inserted before a frozen segmentation backbone. Operating on a three-frame window, LFC fuses consecutive frames via spatio-temporal attention into a single compensated frame, and is trained only on pseudo-labels generated by a differently seeded model, without any expert annotation. Experiments on three polyp video benchmarks show that LFC recovers most of the accuracy degraded by direct frame skipping and, in several settings, reaches the accuracy of frame-by-frame inference, while retaining a 2.05X--2.55X throughput gain. We further evaluate the temporal stability of three backbones spanning CNN, pyramid Transformer, and global-attention Transformer architectures, revealing a spectrum of sensitivity to frame skipping that arises from how locally or globally each architecture aggregates misaligned input. The compensation gain of LFC follows this spectrum: it is largest for high-capacity Transformer backbones, which achieve the best static accuracy but suffer the most under skip-frame inference. LFC thus offers a lightweight, pluggable, and deployment-friendly route to real-time polyp segmentation.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
Topics: Computer science, Artificial intelligence, Computer vision, Segmentation, Frame (networking), Motion compensation
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.31228, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange