Newest papers

Foundation Model Driven Optical Flow

Praroop Chanda, Suryansh Kumar · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26690 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Kami menyelidiki bagaimana representasi visual dan kedalaman pretrained beku mendukung aliran optik padat ketika dikombinasikan dengan jaringan matching yang diawasi. Dibangun pada TTSFlow, makalah ini mempelajari fitur Decoding Dense Prediction Transformer (DPT) dan satu tahap penyempurnaan kasar hingga halus. Metode kami menggabungkan fitur dari model DINOv2 beku dan Deep Anything V2 dan memperkirakan aliran optik melalui pipa matching berbasis transformator yang dapat dilatih. Setelah pelatihan aliran optik pada FlyingChairs dan FlyingThings3D, ia mencapai kesalahan titik akhir 2.37 (EPE) pada pemisahan pelatihan Sintel, Pass Final, dibandingkan dengan 2.52 untuk konfigurasi FlowSeek (T) yang dilaporkan di sini. Masuk RoCo-Spring 2026 kami, RoCo

Machine translation (Meta M2M100); the English original follows.

Abstract

We investigate how frozen pretrained visual and depth representations support dense optical flow when coupled with a supervised matching network. Building on TTSFlow, this paper studies Dense Prediction Transformer (DPT) feature decoding and one coarse-to-fine refinement stage. Our method fuses features from frozen DINOv2 and Depth Anything V2 models and estimates optical flow through a trainable transformer-based matching pipeline. After optical-flow training on FlyingChairs and FlyingThings3D, it achieves 2.37 endpoint error (EPE) on the Sintel training split, Final pass, compared with 2.52 for the FlowSeek (T) configuration reported here. Our RoCo-Spring 2026 entry, RoCo-29.VSAI, obtains a robustness-balanced score of 1.1038, Spring EPE of 1.411, and RobustSpring proxy EPE of 4.739, with a lower balanced score than seven of the twelve listed reference baselines. These results characterize the complete pipeline, which reuses frozen foundation backbones and performs one refinement stage. Code is available at: \url{https://github.com/Visual-and-Spatial-AI-Lab/FFlow}

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Optical flow, Transformer, Matching (statistics), Artificial intelligence, Code (set theory)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange