The 7th AI City Challenge

Naphade, Milind; Wang, Shuo; Anastasiu, David C.; Tang, Zheng; Chang, Ming-Ching; Yao, Yue; Zheng, Liang; Rahman, Mohammed Shaiqur; Arya, Meenakshi S.; Sharma, Anuj; Feng, Qi; Ablavsky, Vitaly; Sclaroff, Stan; Chakraborty, Pranamesh; Prajapati, Sanjita; Li, Alice; Li, Shangru; Kunadharaju, Krishna; Jiang, Shenxin; Chellappa, Rama

doi:10.1109/cvprw59228.2023.00586

Cited by 36 publications

(7 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike the existing MOT benchmark datasets, we consider indoor sequences where targets reappear many times in the scene. In this work, we tested two indoor datasets: 1) own dataset, 2) Multi-camera people tracking dataset in AI city challenge 2023 [51].…”

Section: Datasets and Evaluation Methodology A Datasetsmentioning

confidence: 99%

“…Despite excluding scenarios involving people reappearing, our proposed method achieves a 13% improvement in IDF 1 score over its baseline framework, DeepSORT [5] while maintaining MOTA and MOTP TABLE 5. Performance comparisons in the multi-camera people tracking dataset (AI city challenge 2023 [51]). A baseline of the proposed methods is DeepSORT [5].…”

Section: A Performance Comparisons With Other Methodsmentioning

confidence: 99%

“…For further comparison between methods, we considered benchmark multi-object tracking dataset [51]. As discussed in Section IV-A2, it has 28 indoor for validation, each containing numerous reappearing tracks.…”

Section: A Performance Comparisons With Other Methodsmentioning

confidence: 99%

“…The Multi-camera People Tracking Dataset (Track 1) from the AI City Challenge 2023 [51] is a large-scale and synthetic indoor people tracking. The dataset consists of 22 sequences, each containing 5-6 clips captured by different multiple cameras, as depicted in Figure 8.…”

Section: ) Ai City Challenge 2023 -Multi-camera People Trackingmentioning

confidence: 99%

See 3 more Smart Citations

Rethinking Multi-Object Tracking Based on Re-Identification and Appearance Model Management

Cho

Kim

2023

IEEE Access

View full text Add to dashboard Cite

Conventional multi-object tracking (MOT) methods often fail to track objects that repeatedly disappear and reappear in the scene due to occlusion or moving off the screen. Previous studies have used motion and appearance information to address these issues but have limitations in dealing with complex scenarios. To overcome these limitations, we propose a post-hoc framework for MOT, a so-called rethinking MOT. The proposed framework efficiently combines MOT and re-identification methods to handle tracking failures in conventional MOT methods. To this end, we proposed a track reliability checking method to determine unreliable tracks and performed re-identification to correct incorrect tracking results. For reidentification, we proposed appearance aspect-ratio matching that is practical and effective. In addition, we designed an online appearance model management for accurate and efficient re-identification. The proposed framework does not require any offline training stages, such as feature or metric learning, but it is very practical for comparing appearances in real time. In addition, it is a flexible and pluggable framework that can adopt any baseline methods for MOT and Re-id. To validate the proposed methods, we collected 12 new challenging indoor datasets. Unlike the existing benchmark MOT datasets, we consider complex indoor sequences where targets reappear many times in the scene. The experimental results reveal that the proposed methods are promising for MOT under various challenging sequences and outperform state-ofthe-art methods. Compared to baseline of the proposed rethinking MOT, the IDF 1 score was increased by 36.9% in own dataset and 50.4% in the benchmark dataset.

show abstract

Section: Datasets and Evaluation Methodology A Datasetsmentioning

confidence: 99%

Section: A Performance Comparisons With Other Methodsmentioning

confidence: 99%

Section: A Performance Comparisons With Other Methodsmentioning

confidence: 99%

Section: ) Ai City Challenge 2023 -Multi-camera People Trackingmentioning

confidence: 99%

See 2 more Smart Citations

Rethinking Multi-Object Tracking Based on Re-Identification and Appearance Model Management

Cho

Kim

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Visual traffic monitoring systems, as an essential component of traffic management and safety, have broad application prospects. These systems [5][6][7][8] utilize cameras and other sensor devices to capture real-time traffic scenes and analyze and process images or videos using computer vision algorithms to extract traffic information and key data. Visual traffic monitoring systems consist of functions such as vehicle detection, vehicle tracking, license plate recognition, and behavior analysis, which are used to extract traffic information and features.…”

Section: Introductionmentioning

confidence: 99%

YOLOv7-3D: A Monocular 3D Traffic Object Detection Method from a Roadside Perspective

Ye,

Zhang,

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Current autonomous driving systems predominantly focus on 3D object perception from the vehicle’s perspective. However, the single-camera 3D object detection algorithm in the roadside monitoring scenario provides stereo perception of traffic objects, offering more accurate collection and analysis of traffic information to ensure reliable support for urban traffic safety. In this paper, we propose the YOLOv7-3D algorithm specifically designed for single-camera 3D object detection from a roadside viewpoint. Our approach utilizes various information, including 2D bounding boxes, projected corner keypoints, and offset vectors relative to the center of the 2D bounding boxes, to enhance the accuracy of 3D object bounding box detection. Additionally, we introduce a 5-layer feature pyramid network (FPN) structure and a multi-scale spatial attention mechanism to improve feature saliency for objects of different scales, thereby enhancing the detection accuracy of the network. Experimental results demonstrate that our YOLOv7-3D network achieved significantly higher detection accuracy on the Rope3D dataset while reducing computational complexity by 60%.

show abstract