A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
Deep Cuboid Detection: Beyond 2D Bounding Boxes
[article]
2016
arXiv
pre-print
Preserved Fulltext
We localize cuboids with a 2D bounding box, and simultaneously localize the cuboid's corners, effectively producing a 3D interpretation of box-like objects. ...
We present a Deep Cuboid Detector which takes a consumer-quality RGB image of a cluttered scene and localizes all 3D cuboids (box-like objects). ...
In addition to the 2D bounding box enclosing the cuboid, we estimate the location of all 8 vertices. ...
arXiv:1611.10010v1
fatcat:lsff3atibngwrfzhnz7yvcb7ha
ODAM: Object Detection, Association, and Mapping using Posed RGB Video
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
The proposed system relies on a deep learning front-end to detect 3D objects from a given RGB frame and associate them to a global object-based map using a graph neural network (GNN). ...
We present ODAM, a system for 3D Object Detection, Association, and Mapping using posed RGB videos. ...
Extending beyond 3D oriented bounding boxes, several works focus on estimating dense object shapes via shape embedding [45] or CAD model retrieval [27] given posed RGB video. ...
arXiv:2108.10165v1
fatcat:i4fsf2dwdrbfbce4x33szuxgvq
Movement Tube Detection Network Integrating 3D CNN and Object Detection Framework to Detect Fall
2021
Electronics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
The constrained movement tube, which encapsulates the person with a sequence of bounding boxes, has the merits of encapsulating the person closely and avoiding peripheral interference. ...
A movement tube detection network integrating 3D CNN and object detection framework such as SSD is proposed to detect human fall with constrained movement tubes. ...
of the union of bounding box b and bounding box v. ...
doi:10.3390/electronics10080898
fatcat:unysaqvb7rb6zd3uxd2hzthkzy
Reinforced Axial Refinement Network for Monocular 3D Object Detection
[article]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Monocular 3D object detection aims to extract the 3D position and properties of objects from a 2D input image. ...
Conventional approaches sample 3D bounding boxes from the space and infer the relationship between the target object and each of them, however, the probability of effective samples is relatively small ...
(In each group, we also show the 2D Average Precision (2D AP) of 2D detection results, which is the upper bound of AOS). ...
arXiv:2008.13748v1
fatcat:5eorvtrt3fgtzmyhetqkxmbr4a
Human Detection and Segmentation via Multi-view Consensus
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Self-supervised detection and segmentation of foreground objects aims for accuracy without annotated training data. ...
This corresponds to one 2D bounding box for each view. 3. We compute the 3D center and object height that best agree with these 2D bounding boxes in a least-square sense. 4. ...
Because we ultimately aim to perform single-view 2D detection and segmentation, our approach produces bounding boxes and segmentation masks for each individual view. ...
arXiv:2012.05119v2
fatcat:u42bu5v4xnhw7pgwtzjcfvmhu4
IoU Loss for 2D/3D Object Detection
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
However, during the training stage, the common distance loss (, L_1 or L_2) is often adopted as the loss function to minimize the discrepancy between the predicted and ground truth Bounding Box (Bbox). ...
To eliminate the performance gap between training and testing, the IoU loss has been introduced for 2D object detection in and . ...
IoU computation for 2D: axis-aligned and rotated bounding boxes, where the green and red represent the ground truth and predicted bounding box respectively. ...
arXiv:1908.03851v1
fatcat:jvhspf5acnejlgjngyre4xlsbe
A Survey on Deep Learning Based Methods and Datasets for Monocular 3D Object Detection
2021
Electronics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
This study investigates the major breakthroughs and current progress in deep learning-based monocular 3D object detection. ...
Based on this simple sensor modality for practical applications, deep learning-based monocular 3D object detection methods that overcome significant research challenges are categorized and summarized. ...
In practice, coarse cuboids are reported to have sufficient accuracy for determining the 3D bounding boxes of objects by refinement. ...
doi:10.3390/electronics10040517
fatcat:rziqhrkefvelpg3vgb6qxfprte
Human Detection and Segmentation via Multi-view Consensus
2021
2021 IEEE/CVF International Conference on Computer Vision (ICCV)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
This corresponds to one 2D bounding box for each view. 3. We compute the 3D center and object height that best agree with these 2D bounding boxes in a least-square sense. 4. ...
Because we ultimately aim to perform single-view 2D detection and segmentation, our approach produces bounding boxes and segmentation masks for each individual view. ...
doi:10.1109/iccv48922.2021.00285
fatcat:ascyrs54kff6xcptl4lyw2zg5i
Frustum PointNets for 3D Object Detection from RGB-D Data
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Benefited from learning directly in raw point clouds, our method is also able to precisely estimate 3D bounding boxes even under strong occlusion or with very sparse points. ...
Instead of solely relying on 3D proposals, our method leverages both mature 2D object detectors and advanced 3D deep learning for object localization, achieving efficiency as well as high recall for even ...
However, beyond getting 2D bounding boxes or pixel masks, 3D understanding is eagerly in demand in many applications such as autonomous driving and augmented reality (AR). ...
arXiv:1711.08488v2
fatcat:aatdkha3gzgcnoe4rpem6fggaa
Amodal Detection of 3D Objects: Inferring 3D Bounding Boxes from 2D Ones in RGB-Depth Images
2017
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
This paper addresses the problem of amodal perception of 3D object detection. ...
We propose a novel 3D object detection system that simultaneously predicts objects' 3D locations, physical sizes, and orientations in indoor scenes. ...
[26] provides two kinds of 2D ground truth bounding boxes for NYUV2 dataset: 1) projected 2D bounding boxes by fitting visible point clouds, and 2) projected 2D bounding boxes from amodal 3D bounding ...
doi:10.1109/cvpr.2017.50
dblp:conf/cvpr/DengL17
fatcat:b7gxdyda4bauze2esq7fa5be5q
M3D-RPN: Monocular 3D Region Proposal Network for Object Detection
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2019 pre-print arXiv:1907.06038v1 fatcat:qftcp7nxuzhqplykl56brgvmcy
We leverage the geometric relationship of 2D and 3D perspectives, allowing 3D boxes to utilize well-known and powerful convolutional features generated in the image-space. ...
We propose to reduce the gap by reformulating the monocular 3D detection problem as a standalone 3D region proposal network. ...
The point clouds are then sampled using 2D bounding boxes generated from a separate 2D RPN. ...
arXiv:1907.06038v2
fatcat:pydrzhad6jg35ikzive5ze6qui
A General Pipeline for 3D Detection of Vehicles
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
A two-stage convolutional neural network (CNN) is proposed to refine the detected 3D box. This pipeline is tested on the KITTI dataset using two different 2D detection networks. ...
Much of the current methods support 2D vehicle detection. ...
Subsequently, a set of 3D points which fall into the 2D bounding box after projection is selected. With this set, a model fitting algorithm detects the 3D location and 3D bounding box of the vehicle. ...
arXiv:1803.00387v1
fatcat:dluburmvkrh75gecpcjfv7wmny
PerspectiveNet: 3D Object Detection from a Single RGB Image via Perspective Points
[article]
2019
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
We further devise PerspectiveNet, an end-to-end trainable model that simultaneously detects the 2D bounding box, 2D perspective points, and 3D object bounding box for each object from a single RGB image ...
PerspectiveNet yields three unique advantages: (i) 3D object bounding boxes are estimated based on perspective points, bridging the gap between 2D and 3D bounding boxes without the need of category-specific ...
deep learning methods that directly estimates the 3D object bounding boxes from 2D bounding
boxes [54, 34–36]. ...
arXiv:1912.07744v1
fatcat:6v6rh2uiunfwhbsokfnbcws43m
Deep Learning on Monocular Object Pose Detection and Tracking: A Comprehensive Overview
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2021 pre-print arXiv:2105.14291v1 fatcat:cuxf33omr5f23jabf55jkegw3q
Among methods for object pose detection and tracking, deep learning is the most promising one that has shown better performance than others. ...
Therefore, this study presents a comprehensive review of recent progress in object pose detection and tracking that belongs to the deep learning technical route. ...
Specifically, given a series of images, it first detects 2D bounding boxes of objects to generate proposals for each image. ...
arXiv:2105.14291v2
fatcat:2kxd4owthvf7tbcbnlqlqu4r3m
3D Object Proposals using Stereo Imagery for Accurate Object Class Detection
[article]
2017
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
In particular, we employ a convolutional neural net (CNN) that exploits context and depth information to jointly regress to 3D bounding box coordinates and object pose. ...
We then exploit a CNN on top of these proposals to perform object detection. ...
We propose a 3D object proposal method that goes beyond 2D bounding boxes and is capable of generating highquality 3D bounding box proposals. ...
arXiv:1608.07711v2
fatcat:3xhhpw2o7nggnjfmf24w5wy6ty
« Previous
Showing results 1 — 15 out of 677 results