On the Confidence of Stereo Matching in a Deep-Learning Era: A Quantitative Evaluation
- 주제(키워드) Time measurement , Volume measurement , Estimation , Reliability , Q measurement , Pipelines , Machine learning algorithms , Stereo matching , confidence measures , machine learning , deep learning
- 주제(기타) Computer Science, Artificial Intelligence
- 주제(기타) Engineering, Electrical & Electronic
- 설명문(일반) [Poggi, Matteo; Tosi, Fabio; Aleotti, Filippo; Mattoccia, Stefano] Univ Bologna, I-40136 Bologna, Italy; [Kim, Sunok] Korea Univ, Seoul 02841, South Korea; [Kim, Seungryong] Korea Aerosp Univ, Goyang 10540, South Korea; [Min, Dongbo] Ewha Womans Univ, Seoul 03760, South Korea; [Sohn, Kwanghoon] Yonsei Univ, Seoul 03722, South Korea
- 관리정보기술 faculty
- 등재 SCIE, SCOPUS
- OA유형 Green Submitted
- 발행기관 IEEE COMPUTER SOC
- 발행년도 2022
- 총서유형 Journal
- URI http://www.dcollection.net/handler/ewha/000000194549
- 본문언어 영어
- Published As https://doi.org/10.1109/TPAMI.2021.3069706
- PubMed https://pubmed.ncbi.nlm.nih.gov/33798066
초록/요약
Stereo matching is one of the most popular techniques to estimate dense depth maps by finding the disparity between matching pixels on two, synchronized and rectified images. Alongside with the development of more accurate algorithms, the research community focused on finding good strategies to estimate the reliability, i.e., the confidence, of estimated disparity maps. This information proves to be a powerful cue to naively find wrong matches as well as to improve the overall effectiveness of a variety of stereo algorithms according to different strategies. In this paper, we review more than ten years of developments in the field of confidence estimation for stereo matching. We extensively discuss and evaluate existing confidence measures and their variants, from hand-crafted ones to the most recent, state-of-the-art learning based methods. We study the different behaviors of each measure when applied to a pool of different stereo algorithms and, for the first time in literature, when paired with a state-of-the-art deep stereo network. Our experiments, carried out on five different standard datasets, provide a comprehensive overview of the field, highlighting in particular both strengths and limitations of learning-based strategies.
more