Mutual Occlusion Research Articles

A Multi-Scene Roadside Lidar Benchmark towards Digital Twins of Road Intersections

Abstract. In recent years, the evolution of digital twin technology has paved the way for the construction of intelligent holographic intersections. This can be facilitated by utilizing precise point clouds from roadside lidar. With its capability of real-time monitoring, lidar plays a crucial role in enhancing intersection perception, enabling precise detection and tracking of road objects, as well as providing accurate speed estimates. Despite the introduction of few roadside lidar datasets aimed at enhancing supervised learning algorithms, their applicability to intelligent intersection monitoring remains limited. To address this, this paper presents an Intelligent Intersections (Int2sec) dataset, which exhibits several salient features: 1) it encompasses a broad array of urban intersection scenarios accompanied by a substantial quantity of object annotations; 2) the deployment of dual lidar stations facilitates a thorough scanning of scenes, thereby ensuring expansive scene coverage and mitigating the mutual occlusion phenomenon amongst objects; and 3) the dataset not only catalogues the coordinates, dimensions, and orientations of objects but also encompasses additional attributes such as tracking IDs and real-time motion statuses. Furthermore, the paper evaluates the efficacy of various prominent benchmarking networks, providing a critical analysis and prospective for future research.

Enhancing a You Only Look Once-Plated Detector via Auxiliary Textual Coding for Multi-Scale Rotating Remote Sensing Objects in Transportation Monitoring Applications

With the rapid development of intelligent information technologies, remote sensing object detection has played an important role in different field applications. Particularly in recent years, it has attracted widespread attention in assisting with food safety supervision, which still faces troubling issues between oversized parameters and low performance that are challenging to solve. Hence, this article proposes a novel remote sensing detection framework for multi-scale objects with a rotating status and mutual occlusion, defined as EYMR-Net. This proposed approach is established on the YOLO-v7 architecture with a Swin Transformer backbone, which offers multi-scale receptive fields to mine massive features. Then, an enhanced attention module is added to exploit the spatial and dimensional interrelationships among different local characteristics. Subsequently, the effective rotating frame regression mechanism via circular smoothing labels is introduced to the EYMR-Net structure, addressing the problem of horizontal YOLO (You Only Look Once) frames ignoring direction changes. Extensive experiments on DOTA datasets demonstrated the outstanding performance of EYMR-Net, which achieved an impressive mAP0.5 of up to 74.3%. Further ablation experiments verified that our proposed approach obtains a balance between performance and efficiency, which is beneficial for practical remote sensing applications in transportation monitoring and supply chain management.

Detection of Camellia oleifera fruit maturity in orchards based on modified lightweight YOLO

Fruit maturity is the main factor affecting the quality and yield of Camellia oil. At present, accurate and efficient detection of the maturity level of Camellia oleifera fruits in orchards and enabling selective picking for harvesting robots is a crucial issue. Aiming at the challenge of low detection efficiency of Camellia oleifera fruit maturity and complex object detection models that are difficult to deploy to Camellia oleifera selective harvesting robots, a modified lightweight You Only Look Once model (YOLO-LM) based on YOLOv7-tiny is proposed. Specifically, three Criss-Cross Attention (CCA) modules are introduced to the backbone network to reduce the influence of leaves and branches occlusion as well as mutual occlusion between fruits. Then, the Adaptively Spatial Feature Fusion (ASFF) module is applied as the head to solve the limitation of inconsistency across different feature scales on the feature pyramid. Furthermore, the GSConv is introduced to replace the standard convolution of the Neck network to enable the model to simultaneously preserve model accuracy and reduce model complexity. Experiment results demonstrate that the precision, recall, mAP@0.5, parameters, FLOPs, and model size of YOLO-LM reached 93.96 %, 93.32 %, 93.18 %, 10.17 million, 19.46 G, and 19.82 MB, which outperformed YOLOv3, Faster R-CNN (VGG16), Faster R-CNN (ResNet50), YOLOv4, and YOLOv5m. Compared to YOLOv3-tiny, YOLOv4-tiny, YOLOv5s, YOLOv7-tiny, and YOLOv8s, the mAP@0.5 of YOLO-LM increased by 6.24 %, 6.11 %, 2.52 %, 2.45 %, and 0.57 %, respectively. Meanwhile, the parameters and model size of YOLO-LM are slightly higher than that of YOLOv3-tiny, YOLOv4-tiny, YOLOv5s and YOLOv7-tiny, and significantly lower than that of YOLOv8s. In addition, the detection heatmaps of YOLOv7-tiny and YOLO-LM demonstrate these optimization operations contribute to the enhanced performance of maturity detection for Camellia oleifera fruits within natural orchard environments. Overall, the YOLO-LM model can be used for the maturity detection of Camellia oleifera fruits in orchards, which is expected to provide theoretical reference for orchard yield estimation, growth monitoring, planting management optimization and the development of Camellia oleifera selective harvesting robots.

Attention redirection transformer with semantic oriented learning for unbiased scene graph generation

Scene Graph Generation (SGG) plays an important role in scene understanding because all of the objects and relations in an image can be abstracted into a concise topological graph. Due to the complexity of visual scenes, including mutual occlusion between objects and semantic ambiguity, SGG is still a challenging task. Most of the existing models only focused on the context of a single object while contexts provided by paired objects are ignored. In this paper, we propose an Attention Redirection Transformer (ART) to extract pair-level contexts specifically, which is divided into an attention distraction stage and an attention integration stage. In this way, the attention of the model is forced to be redirected, which explores the implicit information in the background. In addition, to incorporate the semantic information of predicates, a Semantic Oriented Learning Module (SOL) is designed, which may assist in getting better textual semantics and also prompts cross-modal information fusion. At last, a self-diversity driven Dual Translation Embedding Module (DTM) is designed, which refines representations of subject and object and makes them distinct. Experimental results on the Visual Genome dataset demonstrate the effectiveness of our proposed method. Moreover, our method outperforms state-of-the-art methods on the mR@K metric. The source codes are released on Github: https://github.com/Nora-Zhang98/ART-SOL.

SPCANet: congested crowd counting via strip pooling combined attention network.

Crowd counting aims to estimate the number and distribution of the population in crowded places, which is an important research direction in object counting. It is widely used in public place management, crowd behavior analysis, and other scenarios, showing its robust practicality. In recent years, crowd-counting technology has been developing rapidly. However, in highly crowded and noisy scenes, the counting effect of most models is still seriously affected by the distortion of view angle, dense occlusion, and inconsistent crowd distribution. Perspective distortion causes crowds to appear in different sizes and shapes in the image, and dense occlusion and inconsistent crowd distributions result in parts of the crowd not being captured completely. This ultimately results in the imperfect capture of spatial information in the model. To solve such problems, we propose a strip pooling combined attention (SPCANet) network model based on normed-deformable convolution (NDConv). We model long-distance dependencies more efficiently by introducing strip pooling. In contrast to traditional square kernel pooling, strip pooling uses long and narrow kernels (1×N or N×1) to deal with dense crowds, mutual occlusion, and overlap. Efficient channel attention (ECA), a mechanism for learning channel attention using a local cross-channel interaction strategy, is also introduced in SPCANet. This module generates channel attention through a fast 1D convolution to reduce model complexity while improving performance as much as possible. Four mainstream datasets, Shanghai Tech Part A, Shanghai Tech Part B, UCF-QNRF, and UCF CC 50, were utilized in extensive experiments, and mean absolute error (MAE) exceeds the baseline, which is 60.9, 7.3, 90.8, and 161.1, validating the effectiveness of SPCANet. Meanwhile, mean squared error (MSE) decreases by 5.7% on average over the four datasets, and the robustness is greatly improved.

A Detection Algorithm for Citrus Huanglongbing Disease Based on an Improved YOLOv8n.

Given the severe impact of Citrus Huanglongbing on orchard production, accurate detection of the disease is crucial in orchard management. In the natural environments, due to factors such as varying light intensities, mutual occlusion of citrus leaves, the extremely small size of Huanglongbing leaves, and the high similarity between Huanglongbing and other citrus diseases, there remains an issue of low detection accuracy when using existing mainstream object detection models for the detection of citrus Huanglongbing. To address this issue, we propose YOLO-EAF (You Only Look Once-Efficient Asymptotic Fusion), an improved model based on YOLOv8n. Firstly, the Efficient Multi-Scale Attention Module with cross-spatial learning (EMA) is integrated into the backbone feature extraction network to enhance the feature extraction and integration capabilities of the model. Secondly, the adaptive spatial feature fusion (ASFF) module is used to enhance the feature fusion ability of different levels of the model so as to improve the generalization ability of the model. Finally, the focal and efficient intersection over union (Focal-EIOU) is utilized as the loss function, which accelerates the convergence process of the model and improves the regression precision and robustness of the model. In order to verify the performance of the YOLO-EAF method, we tested it on the self-built citrus Huanglongbing image dataset. The experimental results showed that YOLO-EAF achieved an 8.4% higher precision than YOLOv8n on the self-built dataset, reaching 82.7%. The F1-score increased by 3.33% to 77.83%, and the mAP (0.5) increased by 3.3% to 84.7%. Through experimental comparisons, the YOLO-EAF model proposed in this paper offers a new technical route for the monitoring and management of Huanglongbing in smart orange orchards.

Real-time citrus variety detection in orchards based on complex scenarios of improved YOLOv7.

Variety detection provides technical support for selecting XinHui citrus for use in the production of XinHui dried tangerine peel. Simultaneously, the mutual occlusion between tree leaves and fruits is one of the challenges in object detection. In order to improve screening efficiency, this paper introduces a YOLO(You Only Look Once)v7-BiGS(BiFormer&GSConv) citrus variety detection method capable of identifying different citrus varieties efficiently. In the YOLOv7-BiGS network model, initially, the BiFormer attention mechanism in the backbone of the YOLOv7-based network strengthens the model's ability to extract citrus' features. In addition, the introduction of the lightweight GSConv convolution in place of the original convolution within the ELAN of the head component effectively streamlines model complexity while maintaining performance integrity. To environment challenge validate the effectiveness of the method, the proposed YOLOv7-BiGS was compared with YOLOv5, YOLOv7, and YOLOv8. In the comparison of YOLOv7-BiGS with YOLOv5, YOLOv7, and YOLOv8, the experimental results show that the precision, mAP and recell of YOLOv7-BiGS are 91%, 93.7% and 87.3% respectively. Notably, compared to baseline methods, the proposed approach exhibited significant enhancements in precision, mAP, and recall by 5.8%, 4.8%, and 5.2%, respectively. To evaluate the efficacy of the YOLOv7-BiGS in addressing challenges posed by complex environmental conditions, we collected occluded images of Xinhui citrus fruits from the Xinhui orchard base for model detection. This research aims to fulfill performance criteria for citrus variety identification, offering vital technical backing for variety detection endeavors.

Target Detection of Diamond Nanostructures Based on Improved YOLOv8 Modeling.

Boron-doped diamond thin films exhibit extensive applications in chemical sensing, in which the performance could be further enhanced by nano-structuring of the surfaces. In order to discover the relationship between diamond nanostructures and properties, this paper is dedicated to deep learning target detection methods. However, great challenges, such as noise, unclear target boundaries, and mutual occlusion between targets, are inevitable during the target detection of nanostructures. To tackle these challenges, DWS-YOLOv8 (DCN + WIoU + SA + YOLOv8n) is introduced to optimize the YOLOv8n model for the detection of diamond nanostructures. A deformable convolutional C2f (DCN_C2f) module is integrated into the backbone network, as is a shuffling attention (SA) mechanism, for adaptively tuning the perceptual field of the network and reducing the effect of noise. Finally, Wise-IoU (WIoU)v3 is utilized as a bounding box regression loss to enhance the model's ability to localize diamond nanostructures. Compared to YOLOv8n, a 9.4% higher detection accuracy is achieved for the present model with reduced computational complexity. Additionally, the enhancement of precision (P), recall (R), mAP@0.5, and mAP@0.5:0.95 is demonstrated, which validates the effectiveness of the present DWS-YOLOv8 method. These methods provide effective support for the subsequent understanding and customization of the properties of surface nanostructures.

High-precision object detection network for automate pear picking

To address the urgent need for agricultural intelligence in the face of increasing agricultural output and a shortage of personnel, this paper proposes a high precision object detection network for automated pear picking tasks. The current object detection method using deep learning does not fully consider the redundant background information of the pear detection scene and the mutual occlusion characteristics of multiple pears, so that the detection accuracy is low and cannot meet the needs of complex automated pear picking detection tasks. The proposed, High-level deformation-perception Network with multi-object search NMS(HDMNet), is based on YOLOv8 and utilizes a high-level Semantic focused attention mechanism module to eliminate irrelevant background information and a deformation-perception feature pyramid network to improve accuracy of long-distance and small scale fruit. A multi-object search non-maximum suppression is also proposed to choose the anchor frame in a combined search method suitable for multiple pears. The experimental results show that the HDMNet parameter amount is as low as 12.9 M, the GFLOPs is 41.1, the mAP is 75.7%, the mAP50 reaches 93.6%, the mAP75 reaches 70.2%, and the FPS reaches 73.0. Compared with other SOTA object detection methods, it has the transcend of real-time detection, low parameter amount, low calculation amount, high precision, and accurate positioning.

Moving toward Automaticity: A Robust Synthetic Occlusion Image Method for High-Throughput Mushroom Cap Phenotype Extraction

Mushroom cap is a key trait in the growth process and its phenotypic parameters are essential for automatic cultivation and smart breeding. However, the edible mushrooms are usually grown densely with mutual occlusion, which is difficult to obtain the phenotypic parameters non-destructively. Although deep learning methods achieve impressive performance with superior generalization capabilities, they require a large amount of ground truth label of the occluded target, which is a challenging task due to the substantial workload. To overcome this obstacle, a novel synthetic cap occlusion image method was proposed for rapidly generating edible mushroom occlusion datasets using raw images, in which the ground truth is obtained from the real world and the occlusion is randomly generated for simulating real scenes. Moreover, variants of amodal instance segmentation models with different backbone were trained and evaluated on our synthetic occlusion image datasets. Finally, an amodal mask-based size estimation method was presented to calculate the width and length of the cap. The experimental results showed that the amodal instance segmentation achieved an AP@[0.5:0.95] of 82%, 93% and 96% on Oudemansiella raphanipes, Agrocybe cylindraceas and Pholiota nameko synthetic cap datasets, respectively, with a size of 1024 × 1024 px, which indicates that our occlusion image synthesis method can effectively simulate the real cap occlusion situation. The size estimation method achieved an R2 of 0.95 and 0.98 between predictive amodal caps and manually labeled caps for the length and width of Agrocybe cylindraceas cap, respectively, which can be applied to obtain the phenotypic parameters of each cap effectively and accurately. These methods not only meet the demand for automatic monitoring of edible mushroom morphology in factories but also provide technical support for intelligent breeding.

An Improved Pedestrian Detection Model Based on YOLOv8 for Dense Scenes

In dense scenes, pedestrians often exhibit a variety of symmetrical features, such as symmetry in body contour, posture, clothing, and appearance. However, pedestrian detection poses challenges due to the mutual occlusion of pedestrians and the small scale of distant pedestrians in the image. To address these challenges, we propose a pedestrian detection algorithm tailored for dense scenarios called YOLO-RAD. In this algorithm, we integrate the concept of receiving field attention (RFA) into the Conv and C2f modules to enhance the feature extraction capability of the network. A self-designed four-layer adaptive spatial feature fusion (ASFF) module is introduced, and shallow pedestrian feature information is added to enhance the multi-scale feature fusion capability. Finally, we introduce a small-target dynamic head structure (DyHead-S) to enhance the capability of detecting small-scale pedestrians. Experimental results on WiderPerson and CrowdHuman, two challenging dense pedestrian datasets, show that compared with YOLOv8n, our YOLO-RAD algorithm has achieved significant improvement in detection performance, and the detection performance of mAP@0.5 has increased by 2.5% and 6%, respectively. The detection performance of mAP@0.5:0.95 was improved by 2.7% and 6.8%, respectively. Therefore, the algorithm can effectively improve the performance of pedestrian detection in dense scenes.

Lightweight environment sensing algorithm for intelligent driving based on improved YOLOv7

AbstractAccurately and quickly detecting obstacles ahead is a prerequisite for intelligent driving. The combined detection scheme of light detection and ranging (LiDAR) and the camera is far more capable of coping with complex road conditions than a single sensor. However, immediately afterward, ensuring the real‐time performance of the sensing algorithms through a significantly increased amount of computation has become a new challenge. For this purpose, the paper introduces an improved dynamic obstacle detection algorithm based on YOLOv7 (You Only Look Once version 7) to overcome the drawbacks of slow and unstable detection of traditional methods. Concretely, Mobilenetv3 supplants the backbone network utilized in the original YOLOv7 architecture, thereby achieving a reduction in computational overhead. It integrates a specialized layer for the detection of small‐scale targets and incorporates a convolutional block attention module to enhance detection efficacy for diminutive obstacles. Furthermore, the framework adopts the Efficient Intersection over Union Loss function, which is specifically designed to mitigate the issue of mutual occlusion among detected objects. On a dataset consisting of 27,362 labelled KITTI data samples, the improved YOLOv7 algorithm achieves 92.6% mean average precision and 82 frames per second, which reduces the Model_size by 85.9% and loses only 1.5% accuracy compared with the traditional YOLOv7 algorithm. In addition, this paper builds a virtual scene to test the improved algorithm and fuses LiDAR and camera data. Experimental results conducted on a test vehicle equipped with a camera and LiDAR sensor demonstrate the effectiveness and significant performance of the method. The improved obstacle detection algorithm proposed in this research can significantly reduce the computational cost of the environment perception task, meet the requirements of real‐world applications, and is crucial for achieving safer and smarter driving.

A group target track‐before‐detect approach using two‐stage strategy with maximum‐likelihood probabilistic data association

AbstractIn tracking scenarios involving groups with dense targets, achieving effective data association is challenging due to mutual occlusion and interference among targets. The complexity of the tracking problem is further exacerbated in low‐observable environments by the increase in false alarm rates. The track‐before‐detect (TBD) is an advanced technology for detecting and tracking low‐observable targets, effectively mitigating data association problems by integrating multi‐frame echo data. However, the existing multi‐target TBD algorithms typically assume that the targets are spatially separated and are not suitable for scenarios involving group targets. A group target maximum‐likelihood probabilistic data association (GT‐ML‐PDA) algorithm, based on the concept of TBD, is proposed to track group targets effectively in low‐observable environments. The proposed algorithm divides group target tracking into two stages: group centre trajectory estimation and individual target trajectory estimation. To enhance the performance of the proposed algorithm, two strategies are suggested: modifying the equivalent measurements and extracting independent measurement sets for individual targets. Simulation results demonstrate that the proposed algorithm is capable of effectively tracking numerous individual targets within a group, even in the presence of heavy clutter.

An Aerial Image Detection Algorithm Based on Improved YOLOv5.

To enhance aerial image detection in complex environments characterized by multiple small targets and mutual occlusion, we propose an aerial target detection algorithm based on an improved version of YOLOv5 in this paper. Firstly, we employ an improved Mosaic algorithm to address redundant boundaries arising from varying image scales and to augment the training sample size, thereby enhancing detection accuracy. Secondly, we integrate the constructed hybrid attention module into the backbone network to enhance the model's capability in extracting pertinent feature information. Subsequently, we incorporate feature fusion layer 7 and P2 fusion into the neck network, leading to a notable enhancement in the model's capability to detect small targets. Finally, we replace the original PAN + FPN network structure with the optimized BiFPN (Bidirectional Feature Pyramid Network) to enable the model to preserve deeper semantic information, thereby enhancing detection capabilities for dense objects. Experimental results indicate a substantial improvement in both the detection accuracy and speed of the enhanced algorithm compared to its original version. It is noteworthy that the enhanced algorithm exhibits a markedly improved detection performance for aerial images, particularly under real-time conditions.

Multi‐target detection and tracking of shallow marine organisms based on improved YOLO v5 and DeepSORT

AbstractIn order to solve the related problems of detection and tracking of shallow marine organisms, this paper designs a YOLO v5 multi‐target detection and tracking algorithm with attention mechanism. When working underwater, the authors usually encounter many difficulties. Different luminosity and complex coral background usually affect the detection of marine organisms. At the same time, the unrestricted movement of marine organisms, the ability to hide behind rocks and algae, and their mutual occlusion while swimming pose additional challenges to this task. Considering the characteristics of shallow marine organisms activity environment, the attention mechanism is added to the feature extraction network of YOLO v5 to reduce redundant information and improve the detection accuracy of shallow marine organism targets in complex environment. The improved algorithm improves the average detection accuracy of marine organisms target detection by 3.2%. Aiming at the problem of shallow marine organisms target tracking, a shallow marine organisms multi‐target tracking algorithm based on improved Deep Simple Online And Realtime Tracking (SORT) is designed. The improved YOLO v5 algorithm is used to replace Faster R‐CNN (Faster Region‐Convolutional Neural Networks) as the detector of DeepSORT tracking algorithm, and the cascade matching strategy is adopted to solve the problem that the target cannot be tracked continuously when it is occluded for a long time. The experimental results show that the shallow marine organisms multi‐target tracking algorithm based on improved DeepSORT reduces the number of id transformation of marine biological target tracking in shallow sea environment, and improves the accuracy of shallow marine organisms multi‐target tracking.

Varifocal occlusion in an optical see-through near-eye display with a single phase-only liquid crystal on silicon

We propose a near-eye display optics system that supports three-dimensional mutual occlusion. By exploiting the polarization-control properties of a phase-only liquid crystal on silicon (LCoS), we achieve real see-through scene masking as well as virtual digital scene imaging using a single LCoS. Dynamic depth control of the real scene mask and virtual digital image is also achieved by using a focus tunable lens (FTL) pair of opposite curvatures. The proposed configuration using a single LCoS and opposite curvature FTL pair enables the self-alignment of the mask and image at an arbitrary depth without distorting the see-through view of the real scene. We verified the feasibility of the proposed optics using two optical benchtop setups: one with two off-the-shelf FTLs for continuous depth control, and the other with a single Pancharatnam–Berry phase-type FTL for the improved form factor.

Amodal Scene Analysis via Holistic Occlusion Relation Inference and Generative Mask Completion

Amodal scene analysis entails interpreting the occlusion relationship among scene elements and inferring the possible shapes of the invisible parts. Existing methods typically frame this task as an extended instance segmentation or a pair-wise object de-occlusion problem. In this work, we propose a new framework, which comprises a Holistic Occlusion Relation Inference (HORI) module followed by an instance-level Generative Mask Completion (GMC) module. Unlike previous approaches, which rely on mask completion results for occlusion reasoning, our HORI module directly predicts an occlusion relation matrix in a single pass. This approach is much more efficient than the pair-wise de-occlusion process and it naturally handles mutual occlusion, a common but often neglected situation. Moreover, we formulate the mask completion task as a generative process and use a diffusion-based GMC module for instance-level mask completion. This improves mask completion quality and provides multiple plausible solutions. We further introduce a large-scale amodal segmentation dataset with high-quality human annotations, including mutual occlusions. Experiments on our dataset and two public benchmarks demonstrate the advantages of our method. code public available at https://github.com/zbwxp/Amodal-AAAI.

FCFormer: fish density estimation and counting in recirculating aquaculture system

In intelligent feeding recirculating aquaculture system, accurately estimating fish population and density is pivotal for management practices and survival rate assessments. However, challenges arise due to mutual occlusion among fish, rapid movement, and complex breeding environments. Traditional object detection methods based on convolutional neural networks (CNN) often fall short in fully addressing the detection demands for fish schools, especially for distant and small targets. In this regard, we introduce a detection framework dubbed FCFormer (Fish Count Transformer). Specifically, the Twins-SVT backbone network is employed first to extract global features of fish schools. To further enhance feature extraction, especially in the fusion of features at different levels, a Bi-FPN aggregation network model with a CAM Count module is incorporated (BiCC). The CAM module aids in focusing more on critical region features, thus rendering feature fusion more cohesive and effective. Furthermore, to precisely predict density maps and elevate the accuracy of fish counting, we devised an adaptive feature fusion regression head: CRMHead. This approach not only optimizes the feature fusion process but also ensures superior counting precision. Experimental results shown that the proposed FCFormer network achieves an accuracy of 97.06%, with a mean absolute error (MAE) of 6.37 and a root mean square error (MSE) of 8.69. Compared to the Twins transformer, there's a 2.02% improvement, outperforming other transformer-based architectures like CCTrans and DM_Count. The presented FCFormer algorithm can be effectively applied to fish density detection in intelligent feeding recirculating aquaculture system, offering valuable input for the development of intelligent breeding management systems.

Exploiting Spatial-Temporal Context for Interacting Hand Reconstruction on Monocular RGB Video

Reconstructing interacting hands from monocular RGB data is a challenging task, as it involves many interfering factors, e.g., self- and mutual occlusion and similar textures. Previous works only leverage information from a single RGB image without modeling their physically plausible relation, which leads to inferior reconstruction results. In this work, we are dedicated to explicitly exploiting spatial-temporal information to achieve better interacting hand reconstruction. On the one hand, we leverage temporal context to complement insufficient information provided by the single frame and design a novel temporal framework with a temporal constraint for interacting hand motion smoothness. On the other hand, we further propose an interpenetration detection module to produce kinetically plausible interacting hands without physical collisions. Extensive experiments are performed to validate the effectiveness of our proposed framework, which achieves new state-of-the-art performance on public benchmarks.

Automatic detection of standing dead trees based on improved YOLOv7 from airborne remote sensing imagery.

Detecting and localizing standing dead trees (SDTs) is crucial for effective forest management and conservation. Due to challenges posed by mountainous terrain and road conditions, conducting a swift and comprehensive survey of SDTs through traditional manual inventory methods is considerably difficult. In recent years, advancements in deep learning and remote sensing technology have facilitated real-time and efficient detection of dead trees. Nevertheless, challenges persist in identifying individual dead trees in airborne remote sensing images, attributed to factors such as small target size, mutual occlusion and complex backgrounds. These aspects collectively contribute to the increased difficulty of detecting dead trees at a single-tree scale. To address this issue, the paper introduces an improved You Only Look Once version 7 (YOLOv7) model that incorporates the Simple Parameter-Free Attention Module (SimAM), an unparameterized attention mechanism. This improvement aims to enhance the network's feature extraction capabilities and increase the model's sensitivity to small target dead trees. To validate the superiority of SimAM_YOLOv7, we compared it with four widely adopted attention mechanisms. Additionally, a method to enhance model robustness is presented, involving the replacement of the Complete Intersection over Union (CIoU) loss in the original YOLOv7 model with the Wise-IoU (WIoU) loss function. Following these, we evaluated detection accuracy using a self-developed dataset of SDTs in forests. The results indicate that the improved YOLOv7 model can effectively identify dead trees in airborne remote sensing images, achieving precision, recall and mAP@0.5 values of 94.31%, 93.13% and 98.03%, respectively. These values are 3.67%, 2.28% and 1.56% higher than those of the original YOLOv7 model. This improvement model provides a convenient solution for forest management.