Efficient Single-Image Depth Estimation on Mobile Devices, Mobile AI &amp; AIM 2022 Challenge: Report

Ignatov, Andrey; Malivenko, Grigory; Timofte, Radu; Treszczotko, Lukasz; Chang, Xin; Ksiazek, Piotr; Łopuszyński, Michał; Maciej, Pioro,; Rudnicki, Rafal; Smyl, Maciej; Ma, Yujie; Li, Zhenyu; Chen, Zehui; Xu, Jialei; Liu, Xianming; Jiang, Junjun; Shi, XueChao; Xu, Difan; Li, Yanan; Wang, Xiaotao; Lei, Lei; Zhang, Ziyu; Wang, Yicheng; Huang, Zilong; Luo, Guozhong; Yu, Gang; Fu, Bin; Li, Jiaqi; Huang, Zihao; Cao, Zhiguo; Conde, Marcos V.; Denis, Sapozhnikov,; Lee, Byeong Hyun; Park, Dong-Won; Hong, Seong-Min; Lee, Joon‐Hee; Lee, Seunggyu; Chun, Se Young

doi:10.1007/978-3-031-25066-8_4

Cited by 5 publications

(1 citation statement)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Supervised methods [35,36,48,53,57] employ various loss functions [10,26,28,36,47,53] to measure the discrepancy between output depth and ground truth. However, models fail to acquire sufficient structural information from sparse annotations of driving scenes.…”

Section: Introductionmentioning

confidence: 99%

Celebrating the 70th Anniversary of School of Mechanical Science and Engineering of Huazhong University of Science & Technology

Wen

2022

IET Collab Intel Manufact

View full text Add to dashboard Cite

Depth estimation aims to predict dense depth maps. In autonomous driving scenes, sparsity of annotations makes the task challenging. Supervised models produce concave objects due to insufficient structural information. They overfit to valid pixels and fail to restore spatial structures. Self-supervised methods are proposed for the problem. Their robustness is limited by pose estimation, leading to erroneous results in natural scenes. In this paper, we propose a supervised framework termed Diffusion-Augmented Depth Prediction (DADP). We leverage the structural characteristics of diffusion model to enforce depth structures of depth models in a plug-and-play manner. An object-guided integrality loss is also proposed to further enhance regional structure integrality by fetching objective information. We evaluate DADP on three driving benchmarks and achieve significant improvements in depth structures and robustness. Our work provides a new perspective on depth estimation with sparse annotations in autonomous driving scenes. CCS CONCEPTS• Computing methodologies → Scene understanding.

show abstract