Search

The process of object tracking involves consistently identifying each instance across frames depending on initial set of object detection(s). Moreover, in multiple object tracking (MOT), the process through tracking-by-detection paradigm consists of performing two common steps consecutively, which are detection and data association. In MOT, it is targeted to associate detections across frames by localizing and identifying all objects of interest. MOT algorithms further keep tracking even the most challenging issues such as revisiting the same view, missing detections, occlusion and temporarily unseen objects, same-appearance objects coexisting in the same frame occur. Hence, re-identification (re-id) appears to be the most powerful tool for assigning the correct identities to each individual instance when aforementioned issues arise. In this work, we propose a similarity-based person re-id framework, called SAT, using a Siamese neural network via shared weights. Once detections are obtained from the backbone SAT applies a Siamese feature extraction model and then we introduce a similarity array for assessing tracklet(s) and detection(s). We examine the performance of SAT on several benchmarks with extensive experiments and statistical tests, where we improve the current state-of-the-art according to commonly used performance metrics with higher accuracy, less ID switches, less false positive and negative rates. [ABSTRACT FROM AUTHOR]

In recent years, the computer vision applications in the robotics have been improved to approach human-like visual perception and scene/context understanding. Following this aspiration, in this study, we explored the possibility of better object manipulation performance by connecting the visual recognition of objects to their physical attributes, such as weight and center of gravity (CoG). To develop and test this idea, an object manipulation platform is built comprising a robotic arm, a depth camera fixed at the top center of the workspace, embedded encoders in the robotic arm mechanism, and microcontrollers for position and force control. Since both the visual recognition and force estimation algorithms use deep learning principles, the test set-up was named as Deep-Table. The objects in the manipulation tests are selected from everyday life and are common to be seen on modern office desktops. The visual object localization and recognition processes are performed from two distinct branches by deep convolutional neural network architectures. We present five of the possible cases, having different levels of information availability on the object weight and CoG in the experiments. The results confirm that using our algorithm, the robotic arm can move different types of objects successfully varying from several grams (empty bottle) to around 250 g (ceramic cup) without failure or tipping. The proposed method also shows that connecting the object recognition with load estimation and contact point further improves the performance characterized by a smoother motion. [ABSTRACT FROM AUTHOR]

With the rapid progress in deep learning and high-performance computing, video-based traffic monitoring systems and analysis of CCTV camera images have witnessed significant advancements. In this paper, we present a novel and automated traffic monitoring system that harnesses the power of robust deep learning models, offering a comprehensive framework for efficient traffic surveillance. Our system introduces several innovative contributions, including a novel approach for lane identity (LaneID) determination and an integrated methodology for comprehensive traffic rule violation detection. Leveraging state-of-the-art algorithms such as HybridNets, YOLOv8, and DeepSORT, carefully selected through comprehensive comparisons, our approach focuses on detecting lane violations by heavy vehicles and considers crucial factors such as vehicle type, speed, and lane positioning to ensure accurate and reliable violation recognition. By integrating LaneIDs, vehicle speed, and orientation, our system achieves more reliable and nuanced violation detection, improving overall efficiency. Through meticulous fine-tuning and training on a custom dataset, our YOLOv8-based vehicle detection achieved a mean average precision (mAP) of 95%, while our speed estimation algorithm, leveraging a combination of pixel per metric (PPM) and frame differences, demonstrated strong performance with an mAP of 93.1%. This fine-tuned and efficient system ensures real-time monitoring, immediate feedback, and accurate lane violation detection, thereby promoting responsible driving behavior. Additionally, we employed the proposed system to detect other traffic violations, achieving an overall accuracy of 94.03%, which also benefits from geometrical information as of the orientation angle. [ABSTRACT FROM AUTHOR]