Vehicle Pose and Shape Estimation Through Multiple Monocular Vision

Ding, Wenhao; Li, Shuaijun; Zhang, Guilin; Lei, Xiangyu; Qian, Huihuan

doi:10.1109/robio.2018.8665155

Cited by 23 publications

(16 citation statements)

References 29 publications

(55 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the keypoint detection task of human body, [15] proposes a structure named hourglass that reduces the dimension and then increases the dimension of an image, and retains the previous information during the dimension increasing process. As this structure has achieved good results, other tasks begin to benefit from it [57], [58], [59]. Recently, a lot of new methods has been proposed to improve the basic structure of hourglass [60], [61], [62].…”

Section: Region Proposal Methodsmentioning

confidence: 99%

Adaptive Multi-Scale Detection of Acoustic Events

Ding

2020

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

The goal of acoustic (or sound) events detection (AED or SED) is to predict the temporal position of target events in given audio segments. This task plays a significant role in safety monitoring, acoustic early warning and other scenarios. However, the deficiency of data and diversity of acoustic event sources make the AED task a tough issue, especially for prevalent data-driven methods. In this paper, we start from analyzing acoustic events according to their time-frequency domain properties, showing that different acoustic events have different time-frequency scale characteristics. Inspired by the analysis, we propose an adaptive multi-scale detection (AdaMD) method. By taking advantage of hourglass neural network and gated recurrent unit (GRU) module, our AdaMD produces multiple predictions at different temporal and frequency resolutions. An adaptive training algorithm is subsequently adopted to combine multi-scale predictions to enhance the overall capability. Experimental results on Detection and Classification of Acoustic Scenes and Events 2017 (DCASE 2017) Task 2, DCASE 2016 Task 3 and DCASE 2017 Task 3 demonstrate that the AdaMD outperforms published state-of-the-art competitors in terms of the metrics of event error rate (ER) and F1-score. The verification experiment on our collected factory mechanical dataset also proves the noiseresistant capability of the AdaMD, providing the possibility for it to be deployed in the complex environment. Index Terms-rare acoustic event detection, adaptive multiscale, hourglass network PLACE PHOTO HERE Liang He Liang HE received the B.S. degree in communication engineering from Civil Aviation

show abstract

Section: Region Proposal Methodsmentioning

confidence: 99%

Adaptive Multi-Scale Detection of Acoustic Events

Ding

2020

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Viewpoints [32] 9.10 N/A 3DVP [24] 11.18 N/A ObjProp3D [33] 17.37 N/A Reconstruct [34] 12.57 N/A Monocular [35] 2.87 / 4.4134 4.73 / 6.21 Ours 3.40 6.10 TABLE III: Results for pose extraction on test set of KITTI dataset evaluated as orientation and translation errors. These errors of the estimated pose with respect to the ground truths are expressed as geodesic distance for the rotation and distance between the centroids of two point sets for the translation error [35] respectively.…”

Section: Methodsmentioning

confidence: 99%

Vehicle pose estimation via regression of semantic points of interest

Lopez

Agudo

Moreno-Noguer

2019

2019 11th International Symposium on Image and Signal Processing and Analysis (ISPA)

View full text Add to dashboard Cite

In this paper we address the problem of extracting vehicle 3D pose from 2D RGB images. An accurate methodology is presented that is capable of locating 3D coordinates of 20 pre-defined semantic vehicle points of interest or keypoints from 2D information. The presented two-step pipeline provides a straightforward way of extracting three-dimensional information from planar images and avoiding also the usage of other sensor that would lead to a more expensive and hard to manage system. The main contribution of this work is the presented dedicated network architectures that are able to locate simultaneously occluded and visible semantic points of interest to convert these 2D points into 3D space in a simple but efficient way. The presented method uses a robust network based on Stack-Hourglass architecture for precise prediction of semantic 2D keypoints from vehicles even if they are occluded. Furthermore, in the second step another dedicated network converts the 2D points into 3D world coordinates and therefore, the 3D pose of the vehicle can be automatically extracted, outperforming stateof-the-art techniques in terms of accuracy.

show abstract

“…Although the stacked hourglass was initially proposed for human pose estimation, multiple works like [45], [46], [6] and [30] have employed them for vehicle pose estimation. One of the first authors, if not the first, to consider adapting stacked hourglass to other problems was Pavlakos et al [45], using a two hourglasses network with intermediate supervision for keypoint localisation.…”

Section: Related Workmentioning

confidence: 99%

“…In [46], Murthy et al proposed a Conditional Random Field (CRF)-Style loss function at the end of each hourglass unit to not only precisely localise each keypoint, but also enforce inter-keypoint distances constraints. In [6], Ding et al built a four-layer modified hourglass network and also used intermediate supervision. In [30], Reddy et al used a stacked hourglass network as initial visible keypoint detector.…”

Section: Related Workmentioning

confidence: 99%

Simple Baseline for Vehicle Pose Estimation: Experimental Validation

et al. 2020

View full text Add to dashboard Cite

Significant progress on human and vehicle pose estimation has been achieved in recent years. The performance of these methods has evolved from poor to remarkable in just a couple of years. This improvement has been obtained from increasingly complex architectures. In this paper, we explore the applicability of simple baseline methods by adding a few deconvolutional layers on a backbone network to estimate heat maps that correspond to the vehicle keypoints. This approach has been proven to be very effective for human pose estimation. The results are analyzed on the PASCAL3D+ dataset, achieving state-of-the-art results. In addition, a set of experiments has been conducted to study current shortcomings in vehicle keypoints labelling, which adversely affect performance. A new strategy for defining vehicle keypoints is presented and validated with our customized dataset with extended keypoints.

show abstract

Vehicle Pose and Shape Estimation Through Multiple Monocular Vision

Cited by 23 publications

References 29 publications

Adaptive Multi-Scale Detection of Acoustic Events

Adaptive Multi-Scale Detection of Acoustic Events

Vehicle pose estimation via regression of semantic points of interest

Simple Baseline for Vehicle Pose Estimation: Experimental Validation

Contact Info

Product

Resources

About