SOTAVerified

AdaBins: Depth Estimation using Adaptive Bins

2020-11-28CVPR 2021Code Available1· sign in to hype

Shariq Farooq Bhat, Ibraheem Alhashim, Peter Wonka

arXivPDF

Code Available — Be the first to reproduce this paper.

Reproduce

Code

Abstract

We address the problem of estimating a high quality dense depth map from a single RGB input image. We start out with a baseline encoder-decoder convolutional neural network architecture and pose the question of how the global processing of information can help improve overall depth estimation. To this end, we propose a transformer-based architecture block that divides the depth range into bins whose center value is estimated adaptively per image. The final depth values are estimated as linear combinations of the bin centers. We call our new building block AdaBins. Our results show a decisive improvement over the state-of-the-art on several popular depth datasets across all metrics. We also validate the effectiveness of the proposed block with an ablation study and provide the code and corresponding pre-trained weights of the new state-of-the-art model.

Tasks

DecoderDepth EstimationMonocular Depth Estimation

Benchmark Results

DatasetModelMetricClaimedVerifiedStatus
NYU-Depth V2AdaBinsRMS0.36—Unverified

Reproductions