Pose-Based View Synthesis for Vehicles: A Perspective Aware Method

Lv, Kai; Sheng, Hao; Zhang, Xiong; Li, Wei; Zheng, Liang

doi:10.1109/tip.2020.2980130

Cited by 34 publications

(11 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the special field of MOT, the application of deep learning is limited to a certain extent compared with object detection [26] and re-identification [27]. MOT is a typical small sample learning problem, which makes it difficult for deep learning methods to give full play to their advantages.…”

Section: B Deep Learning Methodsmentioning

confidence: 99%

Long-Term Tracking With Deep Tracklet Association

Zhang

Sheng

et al. 2020

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

Section: B Deep Learning Methodsmentioning

confidence: 99%

Long-Term Tracking With Deep Tracklet Association

Zhang

Sheng

et al. 2020

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

“…We follow the approach [25] to obtain the rendering data using the 3dsMax software. The GAN-based data is generated by the approach from [28], which can synthesize novel-view vehicles according to the pre-defined poses. More details in the supplement.…”

Section: Comparison With Data Augmentation Methodsmentioning

confidence: 99%

Robust 2D/3D Vehicle Parsing in CVIS

Miao,

Lu,

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

We present a novel approach to robustly detect and perceive vehicles in different camera views as part of a cooperative vehicle-infrastructure system (CVIS). Our formulation is designed for arbitrary camera views and makes no assumptions about intrinsic or extrinsic parameters. First, to deal with multi-view data scarcity, we propose a partassisted novel view synthesis algorithm for data augmentation. We train a part-based texture inpainting network in a self-supervised manner. Then we render the textured model into the background image with the target 6-DoF pose. Second, to handle various camera parameters, we present a new method that produces dense mappings between image pixels and 3D points to perform robust 2D/3D vehicle parsing. Third, we build the first CVIS dataset for benchmarking, which annotates more than 1540 images (14017 instances) from real-world traffic scenarios. We combine these novel algorithms and datasets to develop a robust approach for 2D/3D vehicle parsing for CVIS. In practice, our approach outperforms SOTA methods on 2D detection, instance segmentation, and 6-DoF pose estimation, by 4.5%, 4.3%, and 2.9%, respectively. More details and results are included in the supplement. To facilitate future research, we will release the source code and the dataset on GitHub.

show abstract

“…Unlike other image generation tasks such as human faces or cars [11], human bodies are non‐rigid and have multiple degrees of freedom. Therefore, many researchers make efforts on pose‐guided person image generation.…”

Section: Related Workmentioning

confidence: 99%

CoGAN: Cooperatively trained conditional and unconditional GAN for person image generation

Liu

Sheng

Wang

et al. 2023

IET Image Processing

Self Cite

View full text Add to dashboard Cite

Person image generation aims to synthesize realistic person images that follow the same distribution as the given dataset. Previous attempts can be generally categorized into two classes: conditional GAN and unconditional GAN. The former usually uses pose information as condition to make pose transfer using GAN. The generated person have the same identity as the source person. The latter generates person images from scratch, and the real person images are only used as references for the discriminator. While conditional GAN is widely studied, unconditional GAN is also worth exploring because it can synthesize person image with new identity, which is a useful manner of data augmentation. These two types of generating methods have their different advantages and disadvantages, and sometimes they are complementary. This paper proposes a CoGAN to cooperatively train two types of GANs in an end‐to‐end framework. The two GANs serve different purposes, and can learn from each other during the cooperative learning procedure. The experimental results on public datasets show that the proposed CoGAN improves the performance of both baseline methods, and achieves competitive results compared with state‐of‐the‐art methods.

show abstract

Pose-Based View Synthesis for Vehicles: A Perspective Aware Method

Cited by 34 publications

References 45 publications

Long-Term Tracking With Deep Tracklet Association

Long-Term Tracking With Deep Tracklet Association

Robust 2D/3D Vehicle Parsing in CVIS

CoGAN: Cooperatively trained conditional and unconditional GAN for person image generation

Contact Info

Product

Resources

About