Multi-Target Pose Estimation and Behavior Analysis Based on Symmetric Cascaded AdderNet

6DoF Object Pose Estimation plays important roles in many computer vision tasks such as Augmented Reality and robotics. During the past decade, the community has been witnessing a great improvement of the pose estimation algorithms. In the age of deep learning, the dominating feature point matching methods have been gradually replaced by various one-shot methods or point-predicting methods. However, the good performances are usually obtained on standard benchmark datasets, with even lighting and few occlusions. In this work, we build a new ‘in-the-wild’ image dataset for the task of 6-DoF object pose estimation, termed Wild-6DoF. This data contains fully-annotated images of 5 real-life objects captured in various complex scenes. We compare some state-of-the-art algorithms as well as a deep-learning based point matching method on some commonly-adopted dataset and the newly-proposed dataset. We found that the state-of-the-art methods, which performs well on the well-controlled datasets, could not beat the "old-fashioned" point-matching method. This observation could reminds the computer vision community that the potential of the conventional approach, especially in the real-life applications.

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.

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.

Solid developments have been seen in deep-learning-based pose estimation, but few works have explored performance in dense crowds, such as a classroom scene; furthermore, no specific knowledge is considered in the design of image augmentation for pose estimation. A masked autoencoder was shown to have a non-negligible capability in image reconstruction, where the masking mechanism that randomly drops patches forces the model to build unknown pixels from known pixels. Inspired by this self-supervised learning method, where the restoration of the feature loss induced by the mask is consistent with tackling the occlusion problem in classroom scenarios, we discovered that the transfer performance of the pre-trained weights could be used as a model-based augmentation to overcome the intractable occlusion in classroom pose estimation. In this study, we proposed a top-down pose estimation method that utilized the natural reconstruction capability of missing information of the MAE as an effective occluded image augmentation in a pose estimation task. The difference with the original MAE was that instead of using a 75% random mask ratio, we regarded the keypoint distribution probabilistic heatmap as a reference for masking, which we named Pose Mask. To test the performance of our method in heavily occluded classroom scenes, we collected a new dataset for pose estimation in classroom scenes named Class Pose and conducted many experiments, the results of which showed promising performance.

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.

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 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.

Object detection of small and occluded targets in complex scenarios is a vital yet challenging task in computer vision, with applications in intelligent systems (e.g., kitchen safety supervision). To address limitations of existing models, this study proposes an improved YOLOv10 algorithm with three key innovations. We first introduce a Mosaic-9 data augmentation strategy to enhance small target density in training samples. The traditional PANet in YOLOv10 is replaced by the Bidirectional Feature Pyramid Network (BiFPN), which uses cross-scale bidirectional connections and learnable weights to optimize multi-scale feature fusion. A Squeeze-and-Excitation (SE) channel attention module is integrated into the CSPDarknet backbone to emphasize key feature channels and mitigate background interference for occluded objects. Experiments on a self-constructed dataset (6508 images with multi-scale and occluded targets) show the improved YOLOv10 achieves 69.5% mAP@0.5, a 7.7 percentage point increase over YOLOv10n, while maintaining 12.1 ms inference speed. Ablation studies verify Mosaic-9 enhances small target perception, BiFPN boosts mAP@0.5 by 5.7%, and SE improves occlusion robustness by 4.8%. This work offers a generalizable multi-module optimization framework for YOLO-series models, applicable to various small and occluded target detection tasks, advancing lightweight object detection algorithms and intelligent vision systems.

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.

In the infrared small target images with complex backgrounds, there exist various interferences that share similar characteristics with the target (such as building edges). The accurate detection of small targets is crucial in applications involving infrared search and tracking. However, traditional detection methods based on small target feature detection in a single frame image may result in higher error rates due to insufficient features. Therefore, in this paper, we propose an infrared moving object detection method that integrates spatio-temporal information. To address the limitations of single-frame detection, we introduce a temporal sequence of images to suppress false alarms caused by single-frame detection through analyzing motion features within the sequence. Firstly, based on spatial feature detection, we propose a multi-scale layered contrast feature (MLCF) filtering for preliminary target extraction. Secondly, we utilize the spatio-temporal context (STC) as a feature to track the image sequence point by point, obtaining global motion features. Statistical characteristics are calculated to obtain motion vector data that correspond to abnormal motion, enabling the accurate localization of moving targets. Finally, by combining spatial and temporal features, we determine the precise positions of the targets. The effectiveness of our method is evaluated using a real infrared dataset. Through analysis of the experimental results, our approach demonstrates stronger background suppression capabilities and lower false alarm rates compared to other existing methods. Moreover, our detection rate is similar or even superior to these algorithms, providing further evidence of the efficacy of our algorithm.

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.

The task of 6D object pose estimation from RGB images is an important requirement for autonomous service robots to be able to interact with the real world. In this work, we present a two-step pipeline for estimating the 6 DoF translation and orientation of known objects. Keypoints and Part Affinity Fields (PAFs) are predicted from the input image adopting the OpenPose CNN architecture from human pose estimation. Object poses are then calculated from 2D-3D correspondences between detected and model keypoints via the PnP-RANSAC algorithm. The proposed approach is evaluated on the YCB-Video dataset and achieves accuracy on par with recent methods from the literature. Using PAFs to assemble detected keypoints into object instances proves advantageous over only using heatmaps. Models trained to predict keypoints of a single object class perform significantly better than models trained for several classes.

Accurate pose estimation of object instances is a key aspect in many applications, including augmented reality or robotics. For example, a task of a domestic robot could be to fetch an item from an open drawer. The poses of both, the drawer and the item have to be known by the robot in order to fulfil the task. 6D pose estimation of rigid objects has been addressed with great success in recent years. In large part, this has been due to the advent of consumer-level RGB-D cameras, which provide rich, robust input data. However, the practical use of state-of-the-art pose estimation approaches is limited by the assumption that objects are rigid. In cluttered, domestic environments this assumption does often not hold. Examples are doors, many types of furniture, certain electronic devices and toys. A robot might encounter these items in any state of articulation. This work considers the task of one-shot pose estimation of articulated object instances from an RGB-D image. In particular, we address objects with the topology of a kinematic chain of any length, i.e. objects are composed of a chain of parts interconnected by joints. We restrict joints to either revolute joints with 1 DOF (degrees of freedom) rotational movement or prismatic joints with 1 DOF translational movement. This topology covers a wide range of common objects (see our dataset for examples). However, our approach can easily be expanded to any topology, and to joints with higher degrees of freedom.

In this paper, we propose 3DBodyTex.Pose, a dataset that addresses the task of 3D human pose estimation in-the-wild. Generalization to in-the-wild images remains limited due to the lack of adequate datasets. Existent ones are usually collected in indoor controlled environments where motion capture systems are used to obtain the 3D ground-truth annotations of humans. 3DBodyTex.Pose offers high quality and rich data containing 405 different real subjects in various clothing and poses, and 81k image samples with ground-truth 2D and 3D pose annotations. These images are generated from 200 viewpoints among which 70 challenging extreme viewpoints. This data was created starting from high resolution textured 3D body scans and by incorporating various realistic backgrounds. Retraining a state-of-the-art 3D pose estimation approach using data augmented with 3DBodyTex.Pose showed promising improvement in the overall performance, and a sensible decrease in the per joint position error when testing on challenging viewpoints. The 3DBodyTex.Pose is expected to offer the research community with new possibilities for generalizing 3D pose estimation from monocular in-the-wild images.