# Attention-based Context Aggregation Network for Monocular Depth Estimation

arXiv: https://arxiv.org/abs/1901.10137 · HTML: https://mlsanity.com/paper/1901.10137/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jan 2019.

## Built on by (most cited first, 18 of 18)

- [Diffusion Models for Monocular Depth Estimation: Overcoming Challenging Conditions](https://mlsanity.com/paper/2407.16698/) (2024)
- [OmniControlNet: Dual-stage Integration for Conditional Image Generation](https://mlsanity.com/paper/2406.05871/) (2024)
- [Deep Learning-based Depth Estimation Methods from Monocular Image and Videos: A Comprehensive Survey](https://mlsanity.com/paper/2406.19675/) (2024)
- [GCNDepth: Self-supervised Monocular Depth Estimation based on Graph Convolutional Network](https://mlsanity.com/paper/2112.06782/) (2021)
- [EdgeConv with Attention Module for Monocular Depth Estimation](https://mlsanity.com/paper/2106.08615/) (2021)
- [Visual Attention-based Self-supervised Absolute Depth Estimation using Geometric Priors in Autonomous Driving](https://mlsanity.com/paper/2205.08780/) (2022)
- [S2Net: Accurate Panorama Depth Estimation on Spherical Surface](https://mlsanity.com/paper/2301.05845/) (2023)
- [Monocular Depth Estimation Using Cues Inspired by Biological Vision Systems](https://mlsanity.com/paper/2204.10384/) (2022)
- [U-HRNet: Delving into Improving Semantic Representation of High Resolution Network for Dense Prediction](https://mlsanity.com/paper/2210.07140/) (2022)
- [A Novel 3D-UNet Deep Learning Framework Based on High-Dimensional Bilateral Grid for Edge Consistent Single Image Depth Estimation](https://mlsanity.com/paper/2105.10129/) (2021)
- [CI-Net: Contextual Information for Joint Semantic Segmentation and Depth Estimation](https://mlsanity.com/paper/2107.13800/) (2021)
- [Learning Monocular Depth Estimation via Selective Distillation of Stereo Knowledge](https://mlsanity.com/paper/2205.08668/) (2022)
- [Dynamic Fusion Network For Light Field Depth Estimation](https://mlsanity.com/paper/2104.05969/) (2021)
- [Depth Monocular Estimation with Attention-based Encoder-Decoder Network from Single Image](https://mlsanity.com/paper/2210.13646/) (2022)
- [An intelligent modular real-time vision-based system for environment perception](https://mlsanity.com/paper/2303.16710/) (2023)
- [Endoscopic Depth Estimation Based on Deep Learning: A Survey](https://mlsanity.com/paper/2507.20881/) (2025)
- [Spectral Rectification for Parameter-Efficient Adaptation of Foundation Models in Colonoscopy Depth Estimation](https://mlsanity.com/paper/2603.15374/) (2026)
- [Monocular Depth Estimation from a Single Image: Progress and Opportunities](https://mlsanity.com/paper/2609.01172/) (2026)

## Abstract (arXiv)

Depth estimation is a traditional computer vision task, which plays a crucial role in understanding 3D scene geometry. Recently, deep-convolutional-neural-networks based methods have achieved promising results in the monocular depth estimation field. Specifically, the framework that combines the multi-scale features extracted by the dilated convolution based block (atrous spatial pyramid pooling, ASPP) has gained the significant improvement in the dense labeling task. However, the discretized and predefined dilation rates cannot capture the continuous context information that differs in diverse scenes and easily introduce the grid artifacts in depth estimation. In this paper, we propose an attention-based context aggregation network (ACAN) to tackle these difficulties. Based on the self-attention model, ACAN adaptively learns the task-specific similarities between pixels to model the context information. First, we recast the monocular depth estimation as a dense labeling multi-class classification problem. Then we propose a soft ordinal inference to transform the predicted probabilities to continuous depth values, which can reduce the discretization error (about 1% decrease in RMSE). Second, the proposed ACAN aggregates both the image-level and pixel-level context information for depth estimation, where the former expresses the statistical characteristic of the whole image and the latter extracts the long-range spatial dependencies for each pixel. Third, for further reducing the inconsistency between the RGB image and depth map, we construct an attention loss to minimize their information entropy. We evaluate on public monocular depth-estimation benchmark datasets (including NYU Depth V2, KITTI). The experiments demonstrate the superiority of our proposed ACAN and achieve the competitive results with the state of the arts.