SOURCE-LINKED INTELLIGENCE

FreeFlow: A Bias-free Hierarchical Transformer for Optical Flow Estimation

arXiv · AI, language, vision and robotics · article · Sep 10, 2026 · UTC

Optical flow methods typically rely on task-specific inductive biases, such as correlation volumes, feature warping, and iterative refinement, among others, to reach high accuracy. While effective, such biases constrain the model to predefined heuristics, which can limit its expressivity and lead to more complex pipelines and additional computational cost. We present FreeFlow, a hierarchical transformer built without any flow-specific components, using instead a single feed-forward encoder--decoder. FreeFlow combines three attention variants: window attention for local processing, shifted-wind

Read original source ↗ Open in workspace

recordType

paper

region

Global