ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving

Song, Xibin; Wang, Peng; Zhou, Dingfu; Zhu, Rui; Guan, Chenye; Dai, Yuchao; Su, Hao; Li, Hongdong; Yang, Ruigang

doi:10.48550/arxiv.1811.12222

Cited by 7 publications

(8 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For our training and test, we used an open-source dataset (ApolloCar3D) [ 23 ]. The data include RGB images for vehicle scenes with information about the location and orientation of each vehicle in every scene (see Figure 2 ).…”

Section: Methodsmentioning

confidence: 99%

3D Vehicle Detection and Segmentation Based on EfficientNetB3 and CenterNet Residual Blocks

Kashevnik

Ali

2022

Sensors

View full text Add to dashboard Cite

In this paper, we present a two stages solution to 3D vehicle detection and segmentation. The first stage depends on the combination of EfficientNetB3 architecture with multiparallel residual blocks (inspired by CenterNet architecture) for 3D localization and poses estimation for vehicles on the scene. The second stage takes the output of the first stage as input (cropped car images) to train EfficientNet B3 for the image recognition task. Using predefined 3D Models, we substitute each vehicle on the scene with its match using the rotation matrix and translation vector from the first stage to get the 3D detection bounding boxes and segmentation masks. We trained our models on an open-source dataset (ApolloCar3D). Our method outperforms all published solutions in terms of 6 degrees of freedom error (6 DoF err).

show abstract

Section: Methodsmentioning

confidence: 99%

3D Vehicle Detection and Segmentation Based on EfficientNetB3 and CenterNet Residual Blocks

Kashevnik

Ali

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…1 is a sample in the Sky dataset. While the ApolloCar3D dataset contains 5,277 driving images of 3384 × 2710 resolution and over 60K car instances of 79 car models [7], the Sky dataset is built only upon 3D annotations of their 4,248 images which are chosen randomly. With K is the projection was used in the ApolloCar3D dataset, construct configurations of the problem (names of variables are still remain) for each image in 4,248 chosen images:…”

Section: Sky Datasetmentioning

confidence: 99%

“…For the first case, consider every screen position (u ′ , v ′ ) on L that have q( I, (u ′ , v ′ )) = ∅. Recall that in (7),…”

Section: Appendix a Details About Circles In The Sky Datasetmentioning

confidence: 99%

“…Thus, I will construct a dataset that represents an important special case of the problem, propose a solution for the special case and analyze the solution on the dataset. The dataset, Sky 1 , was constructed on the Apol-loCar3D dataset [7] but was modified to balance practicality and complexity to make it simple enough for analysis. Because of the nature of the ApolloCar3D dataset, lights and material 1 https://github.com/thienannguyen-cv/Sky-dataset appearance are not taken into account.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Solution for a Fundamental Problem of 3D Inference based on 2D Representations

Nguyen

2022

Preprint

View full text Add to dashboard Cite

3D inference from monocular vision using neural networks is an important research area of computer vision. Applications of the research area are various with many solutions being proposed and have shown remarkable performance. Despite the fact that many efforts have been invested, there are still unanswered questions and some of them are fundamental. In this paper, I discuss a fundamental problem and propose a solution for its important special case.

show abstract

“…Recently, Lu et al [171] utilize a novel architecture contains 3D convolutions and RNNs to achieve centimeter-level localization accuracy in different real-world driving scenarios. Song et al [172] release a 3D car instance understanding benchmark for autonomous driving. Banerjee et al [173] utilize sensor fusion to obtain better features.…”

Section: A Typical Application Areasmentioning

confidence: 99%

A Survey of Deep Learning-based Object Detection

Jiao,

Zhang,

Liu

et al. 2019

Preprint

View full text Add to dashboard Cite

Object detection is one of the most important and challenging branches of computer vision, which has been widely applied in peoples life, such as monitoring security, autonomous driving and so on, with the purpose of locating instances of semantic objects of a certain class. With the rapid development of deep learning networks for detection tasks, the performance of object detectors has been greatly improved. In order to understand the main development status of object detection pipeline, thoroughly and deeply, in this survey, we first analyze the methods of existing typical detection models and describe the benchmark datasets. Afterwards and primarily, we provide a comprehensive overview of a variety of object detection methods in a systematic manner, covering the one-stage and two-stage detectors. Moreover, we list the traditional and new applications. Some representative branches of object detection are analyzed as well. Finally, we discuss the architecture of exploiting these object detection methods to build an effective and efficient system and point out a set of development trends to better follow the state-of-the-art algorithms and further research.

show abstract

ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving

Cited by 7 publications

References 48 publications

3D Vehicle Detection and Segmentation Based on EfficientNetB3 and CenterNet Residual Blocks

3D Vehicle Detection and Segmentation Based on EfficientNetB3 and CenterNet Residual Blocks

A Solution for a Fundamental Problem of 3D Inference based on 2D Representations

A Survey of Deep Learning-based Object Detection

Contact Info

Product

Resources

About