Sensor-Guided Optical Flow
This work addresses accuracy enhancement in optical flow estimation for computer vision applications, but it is incremental as it builds on existing networks with external guidance.
The paper tackles the problem of improving optical flow accuracy by guiding a state-of-the-art network with sparse external cues, achieving superior results on known and unseen domains as supported by experimental benchmarks.
This paper proposes a framework to guide an optical flow network with external cues to achieve superior accuracy either on known or unseen domains. Given the availability of sparse yet accurate optical flow hints from an external source, these are injected to modulate the correlation scores computed by a state-of-the-art optical flow network and guide it towards more accurate predictions. Although no real sensor can provide sparse flow hints, we show how these can be obtained by combining depth measurements from active sensors with geometry and hand-crafted optical flow algorithms, leading to accurate enough hints for our purpose. Experimental results with a state-of-the-art flow network on standard benchmarks support the effectiveness of our framework, both in simulated and real conditions.