Pose Mask: A Model-Based Augmentation Method for 2D Pose Estimation in Classroom Scenes Using Surveillance Images

In the tasks of pose estimation and behavior analysis in computer vision, conventional models are often constrained by various factors or complex environments (such as multiple targets, small targets, occluded targets, etc.). To address this problem, this paper proposes a symmetric cascaded additive network (MulAG) to improve the accuracy of posture estimation and behavior analysis in complex environments. MulAG consists of two modules, MulA and MulG. The MulA module is designed based on a cascaded symmetric network structure and incorporates the addition operation. MulA extracts the posture spatial features of the target from a single frame image. And, the MulG module is designed based on three continuous GRUs (gated recurrent unit). Based on the MulA, MulG extracts the posture temporal features from the posture spatial features of the moving target and predicts the posture temporal features of the moving target. The paper firstly demonstrates the feasibility of addition operations in pose estimation tasks by comparing with MobileNet-v3 in ablation experiments. Secondly, on the HiEve and CrowdPose datasets, MulA achieves accuracy of 79.6% and 80.4%, respectively, outperforming the PTM model by 12.0% and 21.2%. And detection speed of MulA achieves the best value at 8.6ms, which is 1 times higher than HDGCN. The result demonstrates the effectiveness of MulA in multi-target pose estimation in complex scenes. Finally, on the HDMB-51 and UCF-101 datasets, MulAG achieves accuracy of 74.8% and 86.3%, respectively, outperforming HDGCN by 9.6% and 9.5%. Compared with SKP and GIST, the fps of MulAG (44.8s-1) is improved by 8.2% and 8.9%. These experiments highlight the generalizability and superiority of MulAG in behavior analysis and pose estimation tasks.

Pose estimation has been a hot topic in the field of machine vision in recent years. Animals exist widely in nature, and the analysis of their shape and movement is important in many fields and industries. In the pose estimation task, to improve the detection accuracy, the existing models often need to consume a lot of computing and memory resources. Therefore, it is a key problem for the pose estimation methods to carry out a lightweight model and reduce the computational overhead on the premise of ensuring model accuracy. In this paper, we focus on the structure of the convolutional neural network in animal pose estimation, construct a lightweight and efficient stacked hourglass network model oriented to optimize the balance of model computation and accuracy, and implement the application algorithm design based on it. Aiming at the problem of large parameters in depthwise convolutional neural networks, a lightweight residual module is proposed, that is, based on the lightweight efficient channel attention improved conditional channel-weighted method (ICCW-Bottle), thereby reducing the weight of the network and obtaining the feature information of different scales. Given the problem that a large amount of feature information is easily lost after the network pooling operation, a lightweight dual-branch fusion module is proposed that fully integrates high-level semantic information and low-level detailed features under the condition of a small number of parameters. Finally, the same as the CC-SSL method: the model is trained jointly using synthetic and real animal datasets, but the CC-SSL method does not take into account the computational power of the model, which consumes a lot of time and memory to run. Through experiments, it is known that compared with the CC-SSL method, the PCK@0.05 of this method is increased by 5.5% on the TigDog dataset. The model in this paper reduces the number of parameters and calculations of the network while ensuring less information loss and model accuracy. The ablation experiment verifies the advancement and effectiveness of the overall network.

Human pose estimation in videos remains a challenge, largely due to the reliance on extensive manual annotation of large datasets, which is expensive and labor-intensive. Furthermore, existing approaches often struggle to capture long-range temporal dependencies and overlook the complementary relationship between temporal pose heatmaps and visual features. To address these limitations, we introduce STDPose, a novel framework that enhances human pose estimation by learning spatiotemporal dynamics in sparsely-labeled videos. STDPose incorporates two key innovations: 1) A novel Dynamic-Aware Mask to capture long-range motion context, allowing for a nuanced understanding of pose changes. 2) A system for encoding and aggregating spatiotemporal representations and motion dynamics to effectively model spatiotemporal relationships, improving the accuracy and robustness of pose estimation. STDPose establishes a new performance benchmark for both video pose propagation (i.e., propagating pose annotations from labeled frames to unlabeled frames) and pose estimation tasks, across three large-scale evaluation datasets. Additionally, utilizing pseudo-labels generated by pose propagation, STDPose achieves competitive performance with only 26.7% labeled data.

In this paper, we propose a method for estimating the camera pose using a classification neural network based on the idea of OCR. Some terrestrial robots can exhibit high maneuverability by freely tilting their upper bodies. When estimating the posture of such robots, posture estimation using IMU and gyroscopic sensors as in the case of drones is affected by the noise generated by the unevenness of the ground, making posture estimation difficult. Pose estimation using deep learning from camera images is one solution to these problems, and various studies have been conducted in the past. However, the accuracy of pose estimation using only inference by deep learning with camera images is extremely poor and is not practical. In order to solve this problem, this paper proposes a classification neural network based on the idea of OCR, which can ensure high inference accuracy in the pose estimation task.

The task of object pose estimation has been a challenge since the early days of computer vision. To estimate the pose (or viewpoint) of an object, people have mostly looked at object intrinsic features, such as shape or appearance. Surprisingly, informative features provided by other, external elements in the scene, have so far mostly been ignored. At the same time, contextual cues have been shown to be of great benefit for related tasks such as object detection or action recognition. In this paper, we explore how information from other objects in the scene can be exploited for pose estimation. In particular, we look at object configurations. We show that, starting from noisy object detections and pose estimates, exploiting the estimated pose and location of other objects in the scene can help to estimate the objects' poses more accurately. We explore both a camera-centered as well as an object-centered representation for relations. Experiments on the challenging KITTI dataset show that object configurations can indeed be used as a complementary cue to appearance-based pose estimation. In addition, object-centered relational representations can also assist object detection.

We present a deployment friendly, fast bottom-up framework for multi-person 3D human pose estimation. We adopt a novel neural representation of multi-person 3D pose which unifies the position of person instances with their corresponding 3D pose representation. This is realized by learning a generative pose embedding which not only ensures plausible 3D pose predictions, but also eliminates the usual keypoint grouping operation as employed in prior bottom-up approaches. Further, we propose a practical deployment paradigm where paired 2D or 3D pose annotations are unavailable. In the absence of any paired supervision, we leverage a frozen network, as a teacher model, which is trained on an auxiliary task of multi-person 2D pose estimation. We cast the learning as a cross-modal alignment problem and propose training objectives to realize a shared latent space between two diverse modalities. We aim to enhance the model’s ability to perform beyond the limiting teacher network by enriching the latent-to-3D pose mapping using artificially synthesized multi-person 3D scene samples. Our approach not only generalizes to in-the-wild images, but also yields a superior trade-off between speed and performance, compared to prior top-down approaches. Our approach also yields state-of-the-art multi-person 3D pose estimation performance among the bottom-up approaches under consistent supervision levels.

Human pose estimation is an important Computer Vision problem, whose goal is to estimate the human body through joints. Currently, methods that employ deep learning techniques excel in the task of 2D human pose estimation. However, the use of 3D poses can bring more accurate and robust results. Since 3D pose labels can only be acquired in restricted scenarios, fully convolutional methods tend to perform poorly on the task. One strategy to solve this problem is to use 2D pose estimators, to estimate 3D poses in two steps using 2D pose inputs. Due to database acquisition constraints, the performance improvement of this strategy can only be observed in controlled environments, therefore domain adaptation techniques can be used to increase the generalization capability of the system by inserting information from synthetic domains. In this work, we propose a novel method called Domain Unified approach, aimed at solving pose misalignment problems on a cross-dataset scenario, through a combination of three modules on top of the pose estimator: pose converter, uncertainty estimator, and domain classifier. Our method led to a 44.1mm (29.24%) error reduction, when training with the SURREAL synthetic dataset and evaluating with Human3.6M over a no-adaption scenario, achieving state-of-the-art performance.

Background: Augmented reality (AR) and artificial intelligence (AI) are highly disruptive technologies that have revolutionised practices in a wide range of domains, including the security sector. Several law enforcement agencies (LEAs) employ AI in their daily operations for forensics and surveillance. AR is also gaining traction in security, particularly with the advent of affordable wearable devices. Equipping police officers with the tools to facilitate an elevated situational awareness (SA) in patrolling and tactical scenarios is expected to improve LEAs' safety and capacity to deliver crucial blows against terrorist and/or criminal threats. Methods: In this paper we present DARLENE, an ecosystem incorporating novel AI techniques for activity recognition and pose estimation tasks, combined with a wearable AR framework for visualization of the inferenced results via dynamic content adaptation according to the wearer's stress level and operational context. The concept has been validated with end-users through co-creation workshops, while the decision-making mechanism for enhancing LEAs' SA has been assessed with experts. Regarding computer vision components, preliminary tests of the instance segmentation method for humans' and objects' detection have been conducted on a subset of videos from the RWF-2000 dataset for violence detection, which have also been used to test a human pose estimation method that has so far exhibited impressive results, constituting the basis of further developments in DARLENE. Results: Evaluation results highlight that target users are positive towards the adoption of the proposed solution in field operations, and that the SA decision-making mechanism produces highly acceptable outcomes. Evaluation of the computer vision components yielded promising results and identified opportunities for improvement. Conclusions: This work provides the context of the DARLENE ecosystem and presents the DARLENE architecture, analyses its individual technologies, and demonstrates preliminary results, which are positive both in terms of technological achievements and user acceptance of the proposed solution.

Vehicle pose estimation with light detection and ranging (LiDAR) is essential in the perception technology of autonomous driving. However, because of incomplete observation measurements and sparsity of the LiDAR point cloud, it is challenging to achieve satisfactory pose extraction based on 3D LiDAR with the existing pose estimation methods. In addition, the demand for real-time performance further increases the difficulty of the pose estimation task. In this paper, we propose a novel vehicle pose estimation method based on the convex hull. The extracted 3D cluster is reduced to the convex hull, reducing the subsequent computation burden while preserving essential contour information. Subsequently, a novel criterion based on the minimum occlusion area is developed for the search-based algorithm, enabling accurate pose estimation. Additionally, this criterion renders the proposed algorithm particularly well-suited for obstacle avoidance. The proposed algorithm is validated on the KITTI dataset and a manually labeled dataset acquired at an industrial park. The results demonstrate that our proposed method can achieve better accuracy than the classical pose estimation method while maintaining real-time speed.

Classroom student counting based on surveillance video is the basis for important tasks such as student behavior analysis, resource optimization, school security, and intelligent management. In recent years, with the development of deep learning, the research of face detection has been greatly promoted, but it still cannot solve the problems of difficult recognition of different poses such as tiny faces in the back row, occlusion of students, and head-down in the classroom. This paper designed an intelligent student counting system based on classroom surveillance video to solve the problem of counting people in classroom scenarios. Also the paper proposed a face detector which mixes feature enhancement modules and background-based modules, using Background information including shoulders, bodies and desks, and search for unobstructed, less-occluded, and head-up targets in multiple video frames in the surveillance video based on the relative stillness of the students in the classroom scene, which finally greatly improves the accuracy of the classroom student counting statistics.

Vehicle pose estimation with LIDAR plays a crucial role in autonomous driving systems. It serves as the fundamental basis for functions such as tracking, path planning, and decision-making. However, the majority of current vehicle pose estimation techniques struggle to produce satisfactory results when faced with incomplete observation measurements, such as L-shaped point cloud clusters without side contours or those including side-view mirrors. In addition, the requirement for real-time results further increases the difficulty of the pose estimation task. In this paper, we present a vehicle Pose Estimation method with Heuristic L-shape fitting and grid-based Particle Filter (PE-HL-PF). We design a geometric shape classifier module to divide clusters into symmetrical and asymmetrical ones according to their shape features. Furthermore, a contour-based heuristic L-shape fitting module is introduced for asymmetrical clusters, and a structure-aware grid-based particle filter is used to estimate the pose of symmetrical clusters. PE-HL-PF first utilizes a heuristic asymmetrical module that selects dominant contours fitting orientation in a heuristic manner, thereby avoiding the need for a complex traversal search. Additionally, a symmetrical module based on particle filtering is incorporated to enhance the stability of orientation estimation. This method achieves significant improvements in both the runtime efficiency and pose estimation accuracy of incomplete point clouds. Compared with state-of-the-art pose estimation methods, our PE-HL-PF demonstrates a notable performance improvement. Our method can estimate the pose of thousands of objects in less than 1 millisecond, a significant improvement over previous methods. The results of experiments performed on the KITTI dataset validate the effectiveness of our approach.

Human detection and pose estimation are essential for understanding human activities in images and videos. Mainstream multi-human pose estimation methods take a top-down approach, where human detection is first performed, then each detected person bounding box is fed into a pose estimation network. This top-down approach suffers from the early commitment of initial detections in crowded scenes and other cases with ambiguities or occlusions, leading to pose estimation failures. We propose the DetPoseNet, an end-to-end multi-human detection and pose estimation framework in a unified three-stage network. Our method consists of a coarse-pose proposal extraction sub-net, a coarse-pose based proposal filtering module, and a multi-scale pose refinement sub-net. The coarse-pose proposal sub-net extracts whole-body bounding boxes and body keypoint proposals in a single shot. The coarse-pose filtering step based on the person and keypoint proposals can effectively rule out unlikely detections, thus improving subsequent processing. The pose refinement sub-net performs cascaded pose estimation on each refined proposal region. Multi-scale supervision and multi-scale regression are used in the pose refinement sub-net to simultaneously strengthen context feature learning. Structure-aware loss and keypoint masking are applied to further improve the pose refinement robustness. Our framework is flexible to accept most existing top-down pose estimators as the role of the pose refinement sub-net in our approach. Experiments on COCO and OCHuman datasets demonstrate the effectiveness of the proposed framework. The proposed method is computationally efficient (5-6x speedup) in estimating multi-person poses with refined bounding boxes in sub-seconds.

We present the first method to perform automatic 3D pose, shape and texture capture of animals from images acquired in-the-wild. In particular, we focus on the problem of capturing 3D information about Grevy's zebras from a collection of images. The Grevy's zebra is one of the most endangered species in Africa, with only a few thousand individuals left. Capturing the shape and pose of these animals can provide biologists and conservationists with information about animal health and behavior. In contrast to research on human pose, shape and texture estimation, training data for endangered species is limited, the animals are in complex natural scenes with occlusion, they are naturally camouflaged, travel in herds, and look similar to each other. To overcome these challenges, we integrate the recent SMAL animal model into a network-based regression pipeline, which we train end-to-end on synthetically generated images with pose, shape, and background variation. Going beyond state-of-the-art methods for human shape and pose estimation, our method learns a shape space for zebras during training. Learning such a shape space from images using only a photometric loss is novel, and the approach can be used to learn shape in other settings with limited 3D supervision. Moreover, we couple 3D pose and shape prediction with the task of texture synthesis, obtaining a full texture map of the animal from a single image. We show that the predicted texture map allows a novel per-instance unsupervised optimization over the network features. This method, SMALST (SMAL with learned Shape and Texture) goes beyond previous work, which assumed manual keypoints and/or segmentation, to regress directly from pixels to 3D animal shape, pose and texture. Code and data are available at https://github.com/silviazuffi/smalst.

We propose and demonstrate an efficient pose estimation method, specifically small tilt angle variation, for visual servo control system design using a monocular camera, eliminating the need for iterative computations. This paper addresses the challenges associated with accurately estimating the relative pose between a chaser, equipped with a monocular camera, and a target that provides visual cues, particularly in scenarios where the captured image shows only minimal variations in the target's tilt angle. Traditional approaches often rely on direct feedback methods from image data or designs for pose estimation and control separately, necessitating computationally expensive iterations for pose estimation or complex implementation of visual servoing. To address these challenges, we adopt an extended Kalman filter, known as a robust nonlinear observer, and propose a non‐iterative approach that integrates motion dynamics models and control inputs into pose estimation from image data. This approach enables efficient integration of pose estimation and control solely relying on a monocular camera, reducing computational costs and improving robustness. Experimental validation using the ball‐and‐plate system demonstrates the effectiveness of our proposed method. In our setup, the camera is deliberately positioned directly above the pose estimation target, creating a scenario with minimal image variation in the tilt angle. Despite this condition, our approach achieves efficient and accurate pose estimation, demonstrating its feasibility for high‐speed pose estimation and control without iterative computations.

In the presented work, innovative low-sampling lightweight architecture is proposed to solve the task of 3D human pose estimation based on 2D key points. The approach introduces specialized trainable pose encodings designed for 3D pose estimation tasks, which are used in conjunction with traditional pose encodings to represent the input data. The architecture of the method includes multilevel feature processing and their adaptive association using a spatial attention mechanism, which allows to enhance relevant features. Experiments conducted on standard test datasets confirmed the effectiveness of the proposed method: a mean joint position error (MPJPE) value of 42.1 was achieved, which exceeds the results of existing approaches.