Location Of Body Parts Research Articles

Body structure processing and attentional patterns in infancy and adulthood.

Infants are sensitive to distortions to the global configurations of bodies by 3.5months of age, suggesting an early onset of body knowledge. It is unclear, however, whether such sensitivity indicates knowledge of the location of specific body parts or solely reflects sensitivity to the overall gestalt of bodies. This study addressed this issue by examining whether, like adults, infants attend to specific locations where body parts have been reorganized. Results show that adults and 5-month-olds, but not 3.5-month-olds, allocated more attention to the body joint areas (e.g., where the arm connects to the shoulder) that were reorganized versus ones that were typical. To examine whether this kind of processing is driven by low-level features, 5-month-olds were tested on images in which the head was removed. Infants no longer exhibited differential scanning of typical versus reorganized bodies. Results suggest that 5-month-olds are sensitive to the location of body parts, thereby demonstrating adult-like response patterns consistent with early expertise in body processing. The contrasting failure of 3.5-month-olds to exhibit sensitivity to the reorganization suggests a developmental change between these ages.

Lizard Body Temperature Acquisition and Lizard Recognition Using Artificial Intelligence.

The acquisition of the body temperature of animals kept in captivity in biology laboratories is crucial for several studies in the field of animal biology. Traditionally, the acquisition process was carried out manually, which does not guarantee much accuracy or consistency in the acquired data and was painful for the animal. The process was then switched to a semi-manual process using a thermal camera, but it still involved manually clicking on each part of the animal's body every 20 s of the video to obtain temperature values, making it a time-consuming, non-automatic, and difficult process. This project aims to automate this acquisition process through the automatic recognition of parts of a lizard's body, reading the temperature in these parts based on a video taken with two cameras simultaneously: an RGB camera and a thermal camera. The first camera detects the location of the lizard's various body parts using artificial intelligence techniques, and the second camera allows reading of the respective temperature of each part. Due to the lack of lizard datasets, either in the biology laboratory or online, a dataset had to be created from scratch, containing the identification of the lizard and six of its body parts. YOLOv5 was used to detect the lizard and its body parts in RGB images, achieving a precision of 90.00% and a recall of 98.80%. After initial calibration, the RGB and thermal camera images are properly localised, making it possible to know the lizard's position, even when the lizard is at the same temperature as its surrounding environment, through a coordinate conversion from the RGB image to the thermal image. The thermal image has a colour temperature scale with the respective maximum and minimum temperature values, which is used to read each pixel of the thermal image, thus allowing the correct temperature to be read in each part of the lizard.

An Orientation-Aware Attention Network for Person Re-Identification

Humans always identify persons through their characteristics, salient attributes, and these attributes’ locations on the body. Most person re-identification methods focus on global and local features corresponding to the former two discriminations, cropping person images into horizontal strips to obtain coarse locations of body parts. However, discriminative clues corresponding to location differences cannot be discovered, so persons with similar appearances are often confused because of their alike components. To address the above problem, we introduce pixel-wise relative positions for the invariance of their orientations in viewpoint changes. To cope with the scale change of relative position, we combine relative positions with self-attention modules that perform on multi-level features. Moreover, in the data augmentation stage, mirrored images are given new labels due to the conversion of the relative position along a horizontal orientation and change in visual chirality. Extensive experiments on four challenging benchmarks demonstrate that the proposed approach shows its superiority and effectiveness in discovering discriminating features.

Automated analysis of animal behavior and its relation to key aspects of the environment reveals new cognitive specializations of neurons

A thorough analysis of animal behavior is essential for examining the relationship between specific neuron activations with the elements of the external environment, behavior, or internal state. Machine learning techniques have made some advancements in automatic segmentation of animal behavior based on data concerning the location of animal body parts [1–3]. At present, these methods cannot achieve the level of segmentation accuracy desired or make correlations between an animal’s behavioral acts and key environmental factors. To address this issue, the authors have created a software package that can extract a variety of behavioral variables from video recordings of animals in experimental settings, enabling mathematical analysis of a behavioral act continuum. The identification of specific aspects of an animal’s anatomy is crucial for extracting a vast array of behavioral variables. In order to accomplish this task, our team employed DeepLabCut, an accessible toolkit for tracking experimental animal behavior that operates on the principle of transfer learning through deep neural networks. We have devised a technique to ascertain the positions of animal body parts in diverse behavioral situations, resulting in a body parts collection meeting two criteria: offering superior responsiveness to small motor movements of the animal and delivering a high percentage of correct body part locations. In scenarios employing camera shooting from above, such a collection encompasses the nose, ears, tail base, body center, forelimbs, hind limbs, and both flanks of the animal’s body. Next, we created software tools to extract and annotate behavioral variables from data on animal kinematics in various cognitive tasks. Our automated system comprises two main scripting modules: CreatePreset and BehaviorAnalyzer. The CreatePreset module interacts with users to select the type of arena geometry, object location, and necessary temporal and spatial parameters for analysis. The script’s result saves as a mat-file for analyzing the behavior of all experiment videos, assuming a constant relative position of the arena and the video camera alongside the experiment’s design. The BehaviorAnalyzer module conducts initial processing on time series data consisting of coordinates of an animal’s body parts. This results in the formation of a kinematogram, which details the kinematics of the body parts. The module then isolates individual behavioral acts of the animal and annotates its behavior based on motivational and environmental factors. Using mutual information-based methods, we analyzed the specialization of hippocampal CA1 neurons in animals as they explored arenas with varying degrees of novelty. Through the analysis, we have identified neurons that exhibit selectivity in relation to specific continuous kinematic parameters governing the posture and trajectory of the animal. These parameters include the animal’s location in the arena space (X and Y coordinates), as well as the speed and angle of rotation of the animal’s head (i.e. absolute orientation in the arena). Neurons specialized in discrete acts of behavior were identified, including rests, locomotions, freezing, rears, and acts of interaction with objects. Furthermore, a selective activation of neurons was found with regard to an additional set of distinct parameters, which combine the animal’s location in the arena and its speed.

Accurate Segmentation of Tilapia Fish Body Parts Based on Deeplabv3+ for Advancing Phenotyping Applications

As an important economic fish resource, germplasm resources and phenotypic measurements of tilapia are of great importance in the direction of culture and genetic improvement. Furthermore, accurate identification and precise localization of tilapia body parts are crucial for enabling key technologies such as automated capture and precise cutting. However, there are some problems in the semantic segmentation of tilapia fish, including the accuracy of target edge segmentation and the ambiguity in segmenting small targets. To improve the accuracy of semantic segmentation of tilapia parts in real farming environments, an improved Deeplabv3+ network model method is proposed for implementing tilapia part segmentation to facilitate phenotypic measurements on tilapia in this paper. The CBAM module is embedded in the encoder, which can improve the accurate identification and localization of tilapia parts by adaptively adjusting the channel weights and spatial weights and better focus on the key features and spatial connections of tilapia body parts. Furthermore, the decoding part of the Deeplabv3+ model is optimized by using SENet, which greatly increases the segmentation accuracy of the network by establishing the interdependence between channels while suppressing useless features. Finally, model performance is tested and compared with the original network and other methods on the tilapia part segmentation dataset. The experimental results show that the segmentation performance of the improved network is better than other networks, such as PSPNet and U-Net, and the IoU values in the head, fins, trunk, and tail of the fish body are 9.78, 2.27, 6.27, and 6.58 percentage points higher than those of the Deeplabv3+ network, respectively. The results validate the effectiveness of our approach in solving the above problems encountered in the semantic segmentation of tilapia parts.

Fish Monitoring from Low-Contrast Underwater Images

A toolset supporting fish detection, orientation, tracking and especially morphological feature estimation with high speed and accuracy, is presented in this paper. It can be exploited in fish farms to automate everyday procedures including size measurement and optimal harvest time estimation, fish health assessment, quantification of feeding needs, etc. It can also be used in an open sea environment to monitor fish size, behavior and the population of various species. An efficient deep learning technique for fish detection is employed and adapted, while methods for fish tracking are also proposed. The fish orientation is classified in order to apply a shape alignment technique that is based on the Ensemble of Regression Trees machine learning method. Shape alignment allows the estimation of fish dimensions (length, height) and the localization of fish body parts of particular interest such as the eyes and gills. The proposed method can estimate the position of 18 landmarks with an accuracy of about 95% from low-contrast underwater images where the fish can be hardly distinguished from its background. Hardware and software acceleration techniques have been applied at the shape alignment process reducing the frame processing latency to less than 0.5 us on a general purpose computer and less than 16 ms on an embedded platform. As a case study, the developed system has been trained and tested with several Mediterranean fish species in the category of seabream. A large public dataset with low-resolution underwater videos and images has also been developed to test the proposed system under worst case conditions.

Attorneys' Questions and Children's Responses Referring to the Nature of Sexual Touch in Child Sexual Abuse Trials.

Eliciting clear descriptions of sexual body parts and abusive touch in child sexual abuse trials is challenging because of children's immaturity and embarrassment. This study examined references to sexual body part knowledge and sexual touch in attorneys' questions and 5- to 10-year-old children's responses (N = 2,247) in 113 child sexual abuse trials. Regardless of children's age, attorneys and children primarily used unclear colloquialisms to refer to sexual body parts. Questions asking children to name sexual body parts elicited more uninformative responses than questions about the function of sexual body parts. In turn, questions about the function of sexual body parts were more likely to increase the specificity of body part identifications than questions about the location of sexual body parts. Attorneys predominantly used option-posing (yes-no and forced choice) questions to ask about sexual body part knowledge, the location of touch, the method or manner of touching, skin-to-skin contact, penetration, and how the touching felt. Generally, wh- questions were no more likely than option-posing questions to elicit uninformative responses, and consistently elicited more child-generated information. The results question the legal assumption that children's uninformative responses when testifying about sexual abuse should be overcome by asking option-posing questions.

CowXNet: An automated cow estrus detection system

Estrus detection is essential for dairy farms to take cows for artificial insemination promptly. Conventional approaches for detecting estrus cows use electronic devices attached to cows to gather data for software analysis. However, electronic devices can be costly and make a cow moody and uncomfortable while moving. In a common approach, observers detect estrus cows by observing their behaviors. However, continuous observation can easily lead to errors due to the observer’s fatigue. Therefore, we designed CowXNet, an automatic estrus detection system for cows, to assist farmers to detect estrus cows. CowXNet requires only a camera attached in a pen and a computer to analyze recorded videos. CowXNet analyzes the estrus behaviors of each cow in a pen and helps farmers to identify estrus cows. To develop and evaluate CowXNet efficiently and effectively, we collected data from Chokchai Farm, the biggest dairy farm in Asia (14.65483∘N, 101.34853∘E). CowXNet has four modules: (i) cow detection uses YOLOv4 to detect cows in recorded videos; (ii) body part detection uses a convolutional neural network to estimate locations of body parts of detected cows; (iii) estrus behavior detection uses body part coordinates to extract a set of discriminative features, and a classification algorithm to detect estrus behaviors, and (iv) behavior analysis module displays estrus behavior for analysis purposes. We evaluated CowXNet for two instances: module-independent evaluation and end-to-end framework evaluation. Overall, CowXNet was promising; it correctly detected estrus behavior interval of cows 83% of cases.

Piscicola geometra (Linnaeus, 1761) leech parasitizing two fish species hosts from Greater Zab near Aski-Kalak, Kurdistan-Iraq, morphological and molecular investigations

Samples of two fish species, Cyprinus carpio and Silurus triostegus were collected from Aski-Kalak and inspected for the presence of parasitic leeches from March 2018 to February 2019 with a prevalence of 6.66 and 8.33 and mean intensity 0.10 and 0.11 respectively. Both hosts’ species were regarded as new hosts for this parasite in Iraq since no previous reports in this country. Collected leeches were fixed, preserved, and identified based on morphological and molecular techniques. For molecular identification, amplification and sequencing of 18S rDNA, with forward primer C1, (ACCCGCTGAACTTTAAGCAT, position 25), and reverse primer C3, (CTCTTCAGAGTACTTTTCAAC, position 390) were studied, while morphological identification based on well descried, and distinguished by focusing on color patterns, number vis arrangement of eyes, body annulations, gonopores shape and location, shape, location and size of body parts, in addition to some species-specific characters, criteria’s were documented by photographing and drawing illustrations.

Men’s Lacrosse Injuries in the 2018 International World Championship Play

Background:Limited data exist on injuries in men’s lacrosse at the international level. As lacrosse’s popularity grows rapidly across the globe, health care providers must understand how to treat lacrosse athletes.Purpose:To analyze injury data from the 2018 Men’s World Lacrosse Championship.Study Design:Descriptive epidemiology study.Methods:This study prospectively observed injuries that occurred during the 2018 Men’s World Lacrosse Championship. The medical staff of each team completed injury report forms, and data were categorized into body part injury, type, mechanism, time, and location of injury.Results:Over 11 days, 1019 athletes competed in 170 games, resulting in a total of 7147 athlete-exposures (AEs). A total of 140 injuries were recorded during the tournament for an injury rate of 19.6 per 1000 AEs (95% CI, 16.4-22.7). Overall, there were more contact injuries (n = 99; 70.7%) than noncontact injuries (n = 41; 29.3%) (P < .0001). Contact injuries most commonly affected the upper extremity, while noncontact injuries most commonly affected the lower extremity. Contusions were the most commonly reported injury type in the tournament (n = 41; 29.3%), followed by ligament sprains (n = 32; 22.9%) and muscle strains (n = 21; 15.0%). Although there was no difference between the first and second half of gameplay, the injury rate increased in the latter portion of each half (the first and third quarters vs the second and fourth quarters) (P < .0001). A total of 4 injuries required trips to the hospital.Conclusion:Lacrosse has a unique injury profile, as it includes both overhead and collision activity as well as multidirectional, cutting movements. Understanding common injury patterns may help with treatment and prevention. Fatigue may play a role in injury rate, and future research of within-game and within-tournament fatigue should explore this relationship.

Bidirectional Interaction Network for Person Re-Identification.

Person re-identification (ReID) task aims to retrieve the same person across multiple spatially disjoint camera views. Due to huge image changes caused by various factors such as posture variation and illumination transformation, images of different persons may share the more similar appearances than images of the same one. Learning discriminative representations to distinguish details of different persons is significant for person ReID. Many existing methods learn discriminative representations resorting to a human body part location branch which requires cumbersome expert human annotations or complex network designs. In this article, a novel bidirectional interaction network is proposed to explore discriminative representations for person ReID without any human body part detection. The proposed method regards multiple convolutional features as responses to various body part properties and exploits the inter-layer interaction to mine discriminative representations for person identities. Firstly, an inter-layer bilinear pooling strategy is proposed to feasibly exploit the pairwise feature relations between two convolution layers. Secondly, to explore interaction of multiple layers, an effective bidirectional integration strategy consisting of two different multi-layer interaction processes is designed to aggregate bilinear pooling interaction of multiple convolution layers. The interaction of multiple layers is implemented in a layer-by-layer nesting policy to ensure the two interaction processes are different and complementary. Extensive experiments validate the superiority of the proposed method on four popular person ReID datasets including Market-1501, DukeMTMC-ReID, CUHK03-NP and MSMT17. Specifically, the proposed method achieves a rank-1 accuracy of 95.1% and 88.2% on Market-1501 and DukeMTMC-ReID, respectively.

OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields.

Realtime multi-person 2D pose estimation is a key component in enabling machines to have an understanding of people in images and videos. In this work, we present a realtime approach to detect the 2D pose of multiple people in an image. The proposed method uses a nonparametric representation, which we refer to as Part Affinity Fields (PAFs), to learn to associate body parts with individuals in the image. This bottom-up system achieves high accuracy and realtime performance, regardless of the number of people in the image. In previous work, PAFs and body part location estimation were refined simultaneously across training stages. We demonstrate that a PAF-only refinement rather than both PAF and body part location refinement results in a substantial increase in both runtime performance and accuracy. We also present the first combined body and foot keypoint detector, based on an internal annotated foot dataset that we have publicly released. We show that the combined detector not only reduces the inference time compared to running them sequentially, but also maintains the accuracy of each component individually. This work has culminated in the release of OpenPose, the first open-source realtime system for multi-person 2D pose detection, including body, foot, hand, and facial keypoints.

Voxelwise modeling reveals selectivity for body part identity and location in BOLD fMRI responses to complex naturalistic stimuli

Voxelwise modeling reveals selectivity for body part identity and location in BOLD fMRI responses to complex naturalistic stimuli

Multi-Task Learning With Coarse Priors for Robust Part-Aware Person Re-Identification

Part-level representations are important for robust person re-identification (ReID), but in practice feature quality suffers due to the body part misalignment problem. In this paper, we present a robust, compact, and easy-to-use method called the Multi-task Part-aware Network (MPN), which is designed to extract semantically aligned part-level features from pedestrian images. MPN solves the body part misalignment problem via multi-task learning (MTL) in the training stage. More specifically, it builds one main task (MT) and one auxiliary task (AT) for each body part on the top of the same backbone model. The ATs are equipped with a coarse prior of the body part locations for training images. ATs then transfer the concept of the body parts to the MTs via optimizing the MT parameters to identify part-relevant channels from the backbone model. Concept transfer is accomplished by means of two novel alignment strategies: namely, parameter space alignment via hard parameter sharing and feature space alignment in a class-wise manner. With the aid of the learned high-quality parameters, MTs can independently extract semantically aligned part-level features from relevant channels in the testing stage. MPN has three key advantages: 1) it does not need to conduct body part detection in the inference stage; 2) its model is very compact and efficient for both training and testing; 3) in the training stage, it requires only coarse priors of body part locations, which are easy to obtain. Systematic experiments on four large-scale ReID databases demonstrate that MPN consistently outperforms state-of-the-art approaches by significant margins.

Robust human pose estimation from distorted wide-angle images through iterative search of transformation parameters

Tracking human motion from video sequences is a well-known video surveillance technique, and many commercially available motion capture devices can now recognize human poses using a depth camera. However, depth camera systems are complicated and have limited optical fields of view. To overcome this problem, it is necessary to develop techniques for recognizing human motion in wide-angle images. In this study, we devised a method for tracking human motion that is robust to wide-angle image distortion. To do so, we developed a new multilayered convolutional neural network architecture for estimating the locations of human body parts in images along with associated transformation parameters that can be applied to a distorted wide-angle image on a frame-by-frame basis. The proposed method was applied to distorted wide-angle images, and its robustness was demonstrated via a quantitative evaluation of human joint prediction and a comparative analysis with a commercially available depth camera-based motion capture system.

DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning.

Quantitative behavioral measurements are important for answering questions across scientific disciplines-from neuroscience to ecology. State-of-the-art deep-learning methods offer major advances in data quality and detail by allowing researchers to automatically estimate locations of an animal's body parts directly from images or videos. However, currently available animal pose estimation methods have limitations in speed and robustness. Here, we introduce a new easy-to-use software toolkit, DeepPoseKit, that addresses these problems using an efficient multi-scale deep-learning model, called Stacked DenseNet, and a fast GPU-based peak-detection algorithm for estimating keypoint locations with subpixel precision. These advances improve processing speed >2x with no loss in accuracy compared to currently available methods. We demonstrate the versatility of our methods with multiple challenging animal pose estimation tasks in laboratory and field settings-including groups of interacting individuals. Our work reduces barriers to using advanced tools for measuring behavior and has broad applicability across the behavioral sciences.

Self and Body Part Localization in Virtual Reality: Comparing a Headset and a Large-Screen Immersive Display.

It is currently not fully understood where people precisely locate themselves in their bodies, particularly in virtual reality. To investigate this, we asked participants to point directly at themselves and to several of their body parts with a virtual pointer, in two virtual reality (VR) setups, a VR headset and a large-screen immersive display (LSID). There was a difference in distance error in pointing to body parts depending on VR setup. Participants pointed relatively accurately to many of their body parts (i.e., eyes, nose, chin, shoulders, and waist). However, in both VR setups when pointing to the feet and the knees they pointed too low, and for the top of the head too high (to larger extents in the VR headset). Taking these distortions into account, the locations found for pointing to self were considered in terms of perceived bodies, based on where the participants had pointed to their body parts in the two VR setups. Pointing to self in terms of the perceived body was mostly to the face, the upper followed by the lower, as well as some to the torso regions. There was no significant overall effect of VR condition for pointing to self in terms of the perceived body (but there was a significant effect of VR if only the physical body (as measured) was considered). In a paper-and-pencil task outside of VR, performed by pointing on a picture of a simple body outline (body template task), participants pointed most to the upper torso. Possible explanations for the differences between pointing to self in the VR setups and the body template task are discussed. The main finding of this study is that the VR setup influences where people point to their body parts, but not to themselves, when perceived and not physical body parts are considered.

Multi-Pig Part Detection and Association with a Fully-Convolutional Network.

Computer vision systems have the potential to provide automated, non-invasive monitoring of livestock animals, however, the lack of public datasets with well-defined targets and evaluation metrics presents a significant challenge for researchers. Consequently, existing solutions often focus on achieving task-specific objectives using relatively small, private datasets. This work introduces a new dataset and method for instance-level detection of multiple pigs in group-housed environments. The method uses a single fully-convolutional neural network to detect the location and orientation of each animal, where both body part locations and pairwise associations are represented in the image space. Accompanying this method is a new dataset containing 2000 annotated images with 24,842 individually annotated pigs from 17 different locations. The proposed method achieves over 99% precision and over 96% recall when detecting pigs in environments previously seen by the network during training. To evaluate the robustness of the trained network, it is also tested on environments and lighting conditions unseen in the training set, where it achieves 91% precision and 67% recall. The dataset is publicly available for download.

Separate multisensory integration processes for ownership and localization of body parts

The experiences that body parts are owned and localized in space are two key aspects of body awareness. Although initial work assumed that the perceived location of one’s body part can be used as a behavioral measure to assess the feeling of owning a body part, recent studies call into question the relationship between localization and ownership of body parts. Yet, little is known about the processes underlying these two aspects of body-part awareness. Here, I applied a statistically optimal cue combination paradigm to a perceptual illusion in which ownership over an artificial hand is experienced, and found that variances predicted by a model of optimal cue combination are similar to those observed in localization of the participant’s hand, but systematically diverge from those observed in ownership of the artificial hand. These findings provide strong evidence for separate processes between ownership and localization of body parts, and indicate a need to revise current models of body part ownership. Results from this study suggest that the neural substrates for perceptual identification of one’s body parts—such as body ownership—are distinct from those underlying spatial localization of the body parts, thus implying a functional distinction between “who” and “where” in the processing of body part information.

The effects of instrumental action on perceptual hand maps

Perceiving the external spatial location of body parts using position sense requires that immediate proprioceptive afferent signals be integrated with information about body size and shape. Longo and Haggard (Proc Natl Acad Sci 107:11727-11732, 2010) developed a method to measure perceptual hand maps reflecting this metric information about body size and shape. In this paradigm, participants indicate the perceived location of landmarks on their occluded hand by pointing with a long baton held in their other hand. By comparing the relative location of judgments of different hand landmarks, perceptual hand maps can be constructed and compared to actual hand structure. The maps show large and highly stereotyped distortions. Here, I investigated the potential effect of biases related to active motor control of the hand doing the pointing in these distortions. Participants localized the fingertip and knuckle of each finger on their occluded left hand either by actively pointing with a baton held in their right hand (pointing condition) or by giving verbal commands to an experimenter on how to move the baton (verbal condition). Similar distortions were clearly apparent in both conditions, suggesting that they are not an artifact of motor control biases related to the pointing hand.