Pixel-Accurate Depth Evaluation in Realistic Driving Scenarios
This provides a more accurate evaluation framework for depth estimation in autonomous driving, particularly for assessing robustness in adverse weather conditions.
The authors introduced a high-resolution depth estimation benchmark with angular resolution up to 25 arcseconds to evaluate depth sensing methods in realistic driving scenarios, demonstrating that stereo approaches provide more stable depth estimates than monocular methods and lidar completion in adverse weather conditions.
This work introduces an evaluation benchmark for depth estimation and completion using high-resolution depth measurements with angular resolution of up to 25" (arcsecond), akin to a 50 megapixel camera with per-pixel depth available. Existing datasets, such as the KITTI benchmark, provide only sparse reference measurements with an order of magnitude lower angular resolution - these sparse measurements are treated as ground truth by existing depth estimation methods. We propose an evaluation methodology in four characteristic automotive scenarios recorded in varying weather conditions (day, night, fog, rain). As a result, our benchmark allows us to evaluate the robustness of depth sensing methods in adverse weather and different driving conditions. Using the proposed evaluation data, we demonstrate that current stereo approaches provide significantly more stable depth estimates than monocular methods and lidar completion in adverse weather. Data and code are available at https://github.com/gruberto/PixelAccurateDepthBenchmark.git.