Depth Is All You Need for Monocular 3D Detection

• URL: http://arxiv.org/abs/2210.02493v1
• Date: Wed, 5 Oct 2022 18:12:30 GMT
• Title: Depth Is All You Need for Monocular 3D Detection
• Authors: Dennis Park, Jie Li, Dian Chen, Vitor Guizilini, Adrien Gaidon
• Abstract summary: We propose to align depth representation with the target domain in unsupervised fashions. Our methods leverage commonly available LiDAR or RGB videos during training time to fine-tune the depth representation, which leads to improved 3D detectors.
• Score: 29.403235118234747
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A key contributor to recent progress in 3D detection from single images is monocular depth estimation. Existing methods focus on how to leverage depth explicitly, by generating pseudo-pointclouds or providing attention cues for image features. More recent works leverage depth prediction as a pretraining task and fine-tune the depth representation while training it for 3D detection. However, the adaptation is insufficient and is limited in scale by manual labels. In this work, we propose to further align depth representation with the target domain in unsupervised fashions. Our methods leverage commonly available LiDAR or RGB videos during training time to fine-tune the depth representation, which leads to improved 3D detectors. Especially when using RGB videos, we show that our two-stage training by first generating pseudo-depth labels is critical because of the inconsistency in loss distribution between the two tasks. With either type of reference data, our multi-task learning approach improves over the state of the art on both KITTI and NuScenes, while matching the test-time complexity of its single task sub-network.

Related papers

• MonoCD: Monocular 3D Object Detection with Complementary Depths [9.186673054867866]
  Depth estimation is an essential but challenging subtask of monocular 3D object detection. We propose to increase the complementarity of depths with two novel designs. Experiments on the KITTI benchmark demonstrate that our method achieves state-of-the-art performance without introducing extra data.
  arXiv  Detail & Related papers  (2024-04-04T03:30:49Z)
• Boosting Monocular 3D Object Detection with Object-Centric Auxiliary Depth Supervision [13.593246617391266]
  We propose a method to boost the RGB image-based 3D detector by jointly training the detection network with a depth prediction loss analogous to the depth estimation task. Our novel object-centric depth prediction loss focuses on depth around foreground objects, which is important for 3D object detection. Our depth regression model is further trained to predict the uncertainty of depth to represent the 3D confidence of objects.
  arXiv  Detail & Related papers  (2022-10-29T11:32:28Z)
• Towards Accurate Reconstruction of 3D Scene Shape from A Single Monocular Image [91.71077190961688]
  We propose a two-stage framework that first predicts depth up to an unknown scale and shift from a single monocular image. We then exploits 3D point cloud data to predict the depth shift and the camera's focal length that allow us to recover 3D scene shapes. We test our depth model on nine unseen datasets and achieve state-of-the-art performance on zero-shot evaluation.
  arXiv  Detail & Related papers  (2022-08-28T16:20:14Z)
• MonoDETR: Depth-guided Transformer for Monocular 3D Object Detection [61.89277940084792]
  We introduce the first DETR framework for Monocular DEtection with a depth-guided TRansformer, named MonoDETR. We formulate 3D object candidates as learnable queries and propose a depth-guided decoder to conduct object-scene depth interactions. On KITTI benchmark with monocular images as input, MonoDETR achieves state-of-the-art performance and requires no extra dense depth annotations.
  arXiv  Detail & Related papers  (2022-03-24T19:28:54Z)
• VR3Dense: Voxel Representation Learning for 3D Object Detection and Monocular Dense Depth Reconstruction [0.951828574518325]
  We introduce a method for jointly training 3D object detection and monocular dense depth reconstruction neural networks. It takes as inputs, a LiDAR point-cloud, and a single RGB image during inference and produces object pose predictions as well as a densely reconstructed depth map. While our object detection is trained in a supervised manner, the depth prediction network is trained with both self-supervised and supervised loss functions.
  arXiv  Detail & Related papers  (2021-04-13T04:25:54Z)
• Sparse Auxiliary Networks for Unified Monocular Depth Prediction and Completion [56.85837052421469]
  Estimating scene geometry from data obtained with cost-effective sensors is key for robots and self-driving cars. In this paper, we study the problem of predicting dense depth from a single RGB image with optional sparse measurements from low-cost active depth sensors. We introduce Sparse Networks (SANs), a new module enabling monodepth networks to perform both the tasks of depth prediction and completion.
  arXiv  Detail & Related papers  (2021-03-30T21:22:26Z)
• Learning Joint 2D-3D Representations for Depth Completion [90.62843376586216]
  We design a simple yet effective neural network block that learns to extract joint 2D and 3D features. Specifically, the block consists of two domain-specific sub-networks that apply 2D convolution on image pixels and continuous convolution on 3D points.
  arXiv  Detail & Related papers  (2020-12-22T22:58:29Z)
• Learning to Recover 3D Scene Shape from a Single Image [98.20106822614392]
  We propose a two-stage framework that first predicts depth up to an unknown scale and shift from a single monocular image. We then use 3D point cloud encoders to predict the missing depth shift and focal length that allow us to recover a realistic 3D scene shape.
  arXiv  Detail & Related papers  (2020-12-17T02:35:13Z)
• Guiding Monocular Depth Estimation Using Depth-Attention Volume [38.92495189498365]
  We propose guiding depth estimation to favor planar structures that are ubiquitous especially in indoor environments. Experiments on two popular indoor datasets, NYU-Depth-v2 and ScanNet, show that our method achieves state-of-the-art depth estimation results.
  arXiv  Detail & Related papers  (2020-04-06T15:45:52Z)
• Single Image Depth Estimation Trained via Depth from Defocus Cues [105.67073923825842]
  Estimating depth from a single RGB image is a fundamental task in computer vision. In this work, we rely, instead of different views, on depth from focus cues. We present results that are on par with supervised methods on KITTI and Make3D datasets and outperform unsupervised learning approaches.
  arXiv  Detail & Related papers  (2020-01-14T20:22:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.