Non-Learning Low-Light Stereo Vision
For stereo vision in low-light conditions, this method offers a non-learning alternative that outperforms learning-based approaches on noisy data.
A non-learning stereo framework using Field of Junctions and boundary-aware SGM produces more accurate sparse disparity maps than recent algorithms on benchmark datasets under severe noise.
We present a non-learning stereo framework for disparity estimation from severely noisy images. Using the Field of Junctions (FoJ), it retains coarse visual features stable under severe noise for cost volume construction while discarding fine textures inseparable from photon noise. The resulting structural information guides boundary-aware Semi-Global Matching (SGM) that dynamically adapts smoothness penalties to preserve true disparity discontinuities. The output is a sparse disparity map more accurate than those of recent stereo algorithms over unmasked pixels on widely-used benchmark datasets.