mlsanityData to 14 Oct 2026

On the Metrics for Evaluating Monocular Depth Estimation

arXiv 2302.10007 ↗First posted 20 Feb 2023cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

Monocular Depth Estimation (MDE) is performed to produce 3D information that can be used in downstream tasks such as those related to on-board perception for Autonomous Vehicles (AVs) or driver assistance. Therefore, a relevant arising question is whether the standard metrics for MDE assessment are a good indicator of the accuracy of future MDE-based driving-related perception tasks. We address this question in this paper. In particular, we take the task of 3D object detection on point clouds as a proxy of on-board perception. We train and test state-of-the-art 3D object detectors using 3D point clouds coming from MDE models. We confront the ranking of object detection results with the ranking given by the depth estimation metrics of the MDE models. We conclude that, indeed, MDE evaluation metrics give rise to a ranking of methods that reflects relatively well the 3D object detection results we may expect. Among the different metrics, the absolute relative (abs-rel) error seems to be the best for that purpose.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 33 of its 38 references that are in the map. First posted on arXiv 20 Feb 2023.

Builds on

33 of 38 references.

References · 33

  1. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  4. 2015SSD: Single Shot MultiBox Detectorcited by 3,130
  5. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,715
  6. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  7. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,274
  8. 2019Objects as Pointscited by 1,175
  9. 2020Center-based 3D Object Detection and Trackingcited by 1,144
  10. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 941
  11. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 820
  12. 2017Unsupervised Learning of Depth and Ego-Motion from Videocited by 782
21 more
  1. 2020AdaBins: Depth Estimation using Adaptive Binscited by 601
  2. 2018Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Drivingcited by 457
  3. 20203DSSD: Point-based 3D Single Stage Object Detectorcited by 440
  4. 20193D Packing for Self-Supervised Monocular Depth Estimationcited by 439
  5. 2020Voxel R-CNN: Towards High Performance Voxel-based 3D Object Detectioncited by 410
  6. 2016Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescuecited by 368
  7. 2018GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Posecited by 342
  8. 2019Pseudo-LiDAR++: Accurate Depth for 3D Object Detection in Autonomous Drivingcited by 188
  9. 2015Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fieldscited by 169
  10. 2017Semi-Supervised Deep Learning for Monocular Depth Map Predictioncited by 109
  11. 2019GS3D: An Efficient 3D Object Detection Framework for Autonomous Drivingcited by 106
  12. 2016Estimating Depth from Monocular Images as Classification Using Deep Fully Convolutional Residual Networkscited by 91
  13. 2020Towards Better Generalization: Joint Depth-Pose Learning without PoseNetcited by 85
  14. 2018Structured Attention Guided Convolutional Neural Fields for Monocular Depth Estimationcited by 78
  15. 2018T2Net: Synthetic-to-Realistic Translation for Solving Single-Image Depth Estimation Taskscited by 70
  16. 2019Geometry-Aware Symmetric Domain Adaptation for Monocular Depth Estimationcited by 61
  17. 2018AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimationcited by 54
  18. 2020SharinGAN: Combining Synthetic and Real Data for Unsupervised Geometry Estimationcited by 25
  19. 2021Monocular Depth Estimation through Virtual-world Supervision and Real-world SfM Self-Supervisioncited by 24
  20. 2020On Deep Learning Techniques to Boost Monocular Depth Estimation for Autonomous Navigationcited by 17
  21. 2018Monocular Depth Estimation by Learning from Heterogeneous Datasetscited by 3

Built on by

Cited by year

Citing papers per year: 2023: 1.

1 paper.

  1. 2023ImageManip: Image-based Robotic Manipulation with Affordance-guided Next View Selectioncited by 7