Adaptive Depth-converted-Scale Convolution for Self-supervised Monocular Depth Estimation
2026-04-09 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the problem of estimating depth from a single camera where object sizes change with distance, causing confusion for existing models. They introduce a new method called Depth-converted-Scale Convolution (DcSConv) that adjusts how the model looks at image features depending on object depth and size, rather than just changing the shape of filters. They also propose a way to combine these new features with traditional ones for better results. Their approach can be added to existing models and improves depth estimation accuracy on a standard dataset. Tests show that each part of their method contributes to the overall improvement.
Monocular depth estimationSelf-supervised learningConvolutional neural networks (CNN)Receptive fieldObject scaleDepth ambiguityKITTI benchmarkFeature fusionAblation studySpatial convolution
Authors
Yanbo Gao, Huibin Bai, Huasong Zhou, Xingyu Gao, Shuai Li, Xun Cai, Hui Yuan, Wei Hua, Tian Xie
Abstract
Self-supervised monocular depth estimation (MDE) has received increasing interests in the last few years. The objects in the scene, including the object size and relationship among different objects, are the main clues to extract the scene structure. However, previous works lack the explicit handling of the changing sizes of the object due to the change of its depth. Especially in a monocular video, the size of the same object is continuously changed, resulting in size and depth ambiguity. To address this problem, we propose a Depth-converted-Scale Convolution (DcSConv) enhanced monocular depth estimation framework, by incorporating the prior relationship between the object depth and object scale to extract features from appropriate scales of the convolution receptive field. The proposed DcSConv focuses on the adaptive scale of the convolution filter instead of the local deformation of its shape. It establishes that the scale of the convolution filter matters no less (or even more in the evaluated task) than its local deformation. Moreover, a Depth-converted-Scale aware Fusion (DcS-F) is developed to adaptively fuse the DcSConv features and the conventional convolution features. Our DcSConv enhanced monocular depth estimation framework can be applied on top of existing CNN based methods as a plug-and-play module to enhance the conventional convolution block. Extensive experiments with different baselines have been conducted on the KITTI benchmark and our method achieves the best results with an improvement up to 11.6% in terms of SqRel reduction. Ablation study also validates the effectiveness of each proposed module.