scholarly journals Joint Pedestrian and Body Part Detection via Semantic Relationship Learning

2019 ◽  
Vol 9 (4) ◽  
pp. 752 ◽  
Author(s):  
Junhua Gu ◽  
Chuanxin Lan ◽  
Wenbai Chen ◽  
Hu Han
Keyword(s):  
Pedestrian Detection ◽  
Body Part ◽  
Upper Body ◽  
Whole Body ◽  
Body Parts ◽  
Semantic Relationship ◽  
The Public ◽  
Novel Approach ◽  
In The Wild ◽  
Person Search

While remarkable progress has been made to pedestrian detection in recent years, robust pedestrian detection in the wild e.g., under surveillance scenarios with occlusions, remains a challenging problem. In this paper, we present a novel approach for joint pedestrian and body part detection via semantic relationship learning under unconstrained scenarios. Specifically, we propose a Body Part Indexed Feature (BPIF) representation to encode the semantic relationship between individual body parts (i.e., head, head-shoulder, upper body, and whole body) and highlight per body part features, providing robustness against partial occlusions to the whole body. We also propose an Adaptive Joint Non-Maximum Suppression (AJ-NMS) to replace the original NMS algorithm widely used in object detection, leading to higher precision and recall for detecting overlapped pedestrians. Experimental results on the public-domain CUHK-SYSU Person Search Dataset show that the proposed approach outperforms the state-of-the-art methods for joint pedestrian and body part detection in the wild.

scholarly journals Visual Body Part Representation in the Lateral Occipitotemporal Cortex in Children/Adolescents and Adults

2020 ◽  
Vol 1 (1) ◽  
Author(s):  
Yuko Okamoto ◽  
Ryo Kitada ◽  
Takanori Kochiyama ◽  
Hiroaki Naruse ◽  
Kai Makita ◽  
...  
Keyword(s):  
Univariate Analysis ◽  
Age Groups ◽  
Body Representation ◽  
Body Part ◽  
Autism Spectrum ◽  
Whole Body ◽  
Body Parts ◽  
Body Images ◽  
Adolescents And Adults ◽  
Occipitotemporal Cortex

Abstract The lateral occipitotemporal cortex (LOTC) that responds to human bodies and body parts has been implicated in social development and neurodevelopmental disorders like autism spectrum disorder (ASD). Neuroimaging studies using a representational similarity analysis (RSA) revealed that body representation in the LOTC of typically developing (TD) adults is categorized into 3 clusters: action effector body parts, noneffector body parts, and face parts. However, its organization of younger people (i.e., children and adolescents) and its association with individual traits remain unclear. In this functional MRI study, TD adults and children/adolescents observed photographs of hands, feet, arms, legs, chests, waists, upper/lower faces, the whole body, and chairs. The univariate analysis showed that fewer child/adolescent participants showed left LOTC activation in response to whole-body images (relative to those of chairs) than adult participants. Contrastingly, the RSA on both age groups revealed a comparable body representation with 3 clusters of body parts in the bilateral LOTC. Hence, this result indicates that, although response to whole-body images can differ, LOTC body part representations for children/ adolescents and adults are highly similar. Furthermore, sensory atypicality is associated with spatial LOTC organization, suggesting the importance of this region for understanding individual difference, which is frequently observed in ASD.

scholarly journals Moving in on human motor cortex. Characterizing the relationship between body parts with non-rigid population Response Fields

2020 ◽  
Author(s):  
Wouter Schellekens ◽  
Carlijn Bakker ◽  
Nick F. Ramsey ◽  
Natalia Petridou
Keyword(s):  
Ring Finger ◽  
Body Part ◽  
7 Tesla ◽  
Whole Body ◽  
Body Parts ◽  
Population Response ◽  
Human Motor Cortex ◽  
Neuronal Populations ◽  
Cortical Motor ◽  
The Relationship

AbstractFor cortical motor activity, the relationships between different body part representations is unknown. Through reciprocal body part relationships, functionality of cortical motor areas with respect to whole body motor control can be characterized. In the current study, we investigate the relationship between body part representations within individual neuronal populations in motor cortices, following a 7 Tesla fMRI 18-body-part motor experiment in combination with our newly developed non-rigid population Response Field (pRF) model and graph theory. The non-rigid pRF metrics reveal somatotopic structures in all included motor cortices covering frontal, parietal, medial and insular cortices and that neuronal populations in primary sensorimotor cortex respond to fewer body parts than secondary motor cortices. Reciprocal body part relationships are estimated in terms of uniqueness, clique-formation, and importance. We report unique response profiles for the knee, a clique of body parts surrounding the ring finger, and a central role for the shoulder and wrist. These results reveal associations among body parts from the perspective of the central nervous system, while being in agreement with intuitive notions of body part usage.

Whole-Body Human-to-Humanoid Motion Imitation

2013 ◽  
Vol 479-480 ◽  
pp. 617-621
Author(s):  
Hsien I Lin ◽  
Zan Sheng Chen
Keyword(s):  
Degrees Of Freedom ◽  
Humanoid Robot ◽  
Optimal Solution ◽  
Human Motion ◽  
Body Motion ◽  
Upper Body ◽  
Whole Body ◽  
Nao Robot ◽  
Novel Approach ◽  
Intuitive Method

Human-to-Humanoid motion imitation is an intuitive method to teach a humanoid robot how to act by human demonstration. For example, teaching a robot how to stand is simply showing the robot how a human stands. Much of previous work in motion imitation focuses on either upper-body or lower-body motion imitation. In this paper, we propose a novel approach to imitate human whole-body motion by a humanoid robot. The main problem of the proposed work is how to control robot balance and keep the robot motion as similar as taught human motion simultaneously. Thus, we propose a balance criterion to assess how well the root can balance and use the criterion and a genetic algorithm to search a sub-optimal solution, making the root balanced and its motion similar to human motion. We have validated the proposed work on an Aldebaran Robotics NAO robot with 25 degrees of freedom. The experimental results show that the root can imitate human postures and autonomously keep itself balanced.

How do you smile along a path?

2019 ◽  
Vol 36 (3) ◽  
pp. 343-363 ◽  
Author(s):  
Patricia Irwin
Keyword(s):  
Syntactic Structure ◽  
Body Part ◽  
The Body ◽  
Whole Body ◽  
Body Parts ◽  
Syntactic Category ◽  
Root Structure ◽  
Core Component ◽  
Syntactic Structures ◽  
Root Content

Abstract This paper argues that a core component of root meaning is the distinction between body parts versus the body conceived as a whole. This distinction is shown to be relevant in the acceptability of motion sentences in English with whole-body roots like $\sqrt {\textsc{dance}} $ and body-part roots like $\sqrt {\textsc{smile}} $. In keeping with the assumption that roots lack syntactic category, I argue that verbal roots occur freely in syntactic structures but that some root-structure combinations are degraded (or unacceptable), and that this is due to an incompatibility between conceptual root content and interpreted syntactic structure.

Action Recognition from Pose Signature in Static Image

2016 ◽  
Vol 30 (03) ◽  
pp. 1655010 ◽  
Author(s):  
Yinzhong Qian ◽  
Wenbin Chen ◽  
I-fan Shen
Keyword(s):  
Performance Improvement ◽  
Action Recognition ◽  
State Of The Art ◽  
Body Part ◽  
The Body ◽  
Whole Body ◽  
Body Parts ◽  
Static Image ◽  
Image Dataset ◽  
Part Structure

This paper addresses the problem of action recognition from body pose. Detecting body pose in static image faces great challenges because of pose variability. Our method is based on action-specific hierarchical poselet. We use hierarchical body parts each of which is represented by a set of poselets to demonstrate the pose variability of the body part. Pose signature of a body part is represented by a vector of detection responses of all poselets for the part. In order to suppress detection error and ambiguity we explore to use part-based model (PBM) as detection context. We propose a constrained optimization algorithm for detecting all poselets of each part in context of PBM, which recover neglected pose clue by global optimization. We use a PBM with hierarchical part structure, where body parts have varying granularity from whole body steadily decreasing to limb parts. From the structure we get models with different depth to study saliency of different body parts in action recognition. Pose signature of an action image is composed of pose signature of all the body parts in the PBM, which provides rich discriminate information for our task. We evaluate our algorithm on two datasets. Compared with counterpart methods, pose signature has obvious performance improvement on static image dataset. While using the model trained from static image dataset to label detected action person on video dataset, pose signature achieves state-of-the-art performance.

scholarly journals Dlaczego Bałtowie i Słowianie liczyli pokrewieństwo w kolanach, czyli o etymologii słow. kolěno ‘kolano; ród’ i lit. kẽlis ‘kolano; ród’

LingVaria ◽  
2020 ◽  
pp. 251-264
Author(s):  
Norbert Ostrowski
Keyword(s):  
Knee Joint ◽  
Body Part ◽  
The Body ◽  
Customary Law ◽  
Middle High German ◽  
Body Parts ◽  
Semantic Relationship ◽  
Low German ◽  
Primary Meaning ◽  
Middle Low German

WHY DID BALTS AND SLAVS COUNT KINSHIP IN KNEES, OR THE ETYMOLOGY OF SLAV. kolěno ‘KNEE; TRIBE’ AND LITH. kẽlis ‘KNEE; TRIBE’ OCS. kolěno ‘knee; tribe, generation’ (cf. Polish pokolenie ‘generation’) and Lith. kẽlis ‘knee; joint in a plant; tribe; degree of kinship’ come from old adjectives with possessive suffixes (-ěn- in Slavic and -ija- in Baltic). Their primary meaning was ‘a joint in the body’ (*‘a rotating part of the body’). Both were formed from nouns with the meaning ‘wheel’ (OCS. kolo, kolese ‘wheel’, Old Prussian kelan ‘wheel’, Latvian duceles ‘chaise’; IDE. *kwelh1- ‘to turn, to rotate’). The hypothesis proposed in this paper explains the semantic relationship between Slav. kolěno ‘knee; tribe, generation’ and Proto-Slav. *kel-nŭ- > *čelnŭ- > Slovenian člèn // Serbo-Croatian člȃn ‘joint; ankle, talus’. Assuming that the meaning ‘joint’ was the original one, OCS. kolěno ‘tribe, generation’ and Lith. kẽlis ‘degree in relationship; tribe’ can be interpreted as old terms of customary law in the field of succession. The counting of kinship by enumeration of body parts from the head to the middle fingernail has been preserved in Middle Low German customary law, so-called “Sachsenspiegel”. The hypothesis is supported by numerous parallels: Lith. sąnarys ‘a joint in the body; (OLith.) ‘generation’, Lith. stráipsnis ‘body part; (OLith.) generation’, OPr. streipstan ‘body part; generation’, Middle High German Gelied ‘body part; generation’, and Middle Low German lede ‘a joint in the body; body part; degree of kinship’.

scholarly journals A Multi-Semantic Driver Behavior Recognition Model of Autonomous Vehicles Using Confidence Fusion Mechanism

Actuators ◽  
2021 ◽  
Vol 10 (9) ◽  
pp. 218
Author(s):  
Hongze Ren ◽  
Yage Guo ◽  
Zhonghao Bai ◽  
Xiangyu Cheng
Keyword(s):  
Autonomous Vehicles ◽  
Driver Behavior ◽  
Upper Body ◽  
Body Parts ◽  
Recognition Model ◽  
Semantic Relationships ◽  
The Public ◽  
Object Features ◽  
Traditional Roles

With the rise of autonomous vehicles, drivers are gradually being liberated from the traditional roles behind steering wheels. Driver behavior cognition is significant for improving safety, comfort, and human–vehicle interaction. Existing research mostly analyzes driver behaviors relying on the movements of upper-body parts, which may lead to false positives and missed detections due to the subtle changes among similar behaviors. In this paper, an end-to-end model is proposed to tackle the problem of the accurate classification of similar driver actions in real-time, known as MSRNet. The proposed architecture is made up of two major branches: the action detection network and the object detection network, which can extract spatiotemporal and key-object features, respectively. Then, the confidence fusion mechanism is introduced to aggregate the predictions from both branches based on the semantic relationships between actions and key objects. Experiments implemented on the modified version of the public dataset Drive&Act demonstrate that the MSRNet can recognize 11 different behaviors with 64.18% accuracy and a 20 fps inference time on an 8-frame input clip. Compared to the state-of-the-art action recognition model, our approach obtains higher accuracy, especially for behaviors with similar movements.

scholarly journals An Exploration of Relationships among Thermal Insulation, Area Factor and Air Gap of Male Chinese Ethnic Costumes

Polymers ◽  
2020 ◽  
Vol 12 (6) ◽  
pp. 1302
Author(s):  
Ying Ke ◽  
Faming Wang
Keyword(s):  
Thermal Insulation ◽  
Minority Groups ◽  
Body Part ◽  
Air Gap ◽  
Whole Body ◽  
Gap Size ◽  
Body Parts ◽  
Air Volume ◽  
Local Thermal Insulation ◽  
Local Thermal Comfort

The present study investigated total and local thermal insulations of 39 sets of male Chinese ethnic costumes. Total and local clothing area factor, air gap size and air volume were determined by a 3D body scanner. Relationships between thermal insulation and air gap for the whole body, as well as local body parts, were explored. Correlations of both the total and local clothing area factor with the intrinsic insulation were also developed. Results demonstrated that the clothing total thermal insulation first increased with the increasing air gap size/air volume, followed by a decrease when the air gap size/air volume exceeded 37.8 mm/55.8 dm3. Similarly, it was also found that parabolic relationships widely existed between the local thermal insulation and local air gap at each body part. Our research findings provide a comprehensive database for predicting both global and local thermal comfort of male Chinese minority groups.

scholarly journals Successful Use of Geochemical Tools to Trace the Geographic Origin of Long-Snouted Seahorse Hippocampus guttulatus Raised in Captivity

Animals ◽  
2021 ◽  
Vol 11 (6) ◽  
pp. 1534
Author(s):  
Ana Elisa Cabral ◽  
Fernando Ricardo ◽  
Carla Patinha ◽  
Eduardo Ferreira da Silva ◽  
Miguel Correia ◽  
...  
Keyword(s):  
Geographic Origin ◽  
Body Part ◽  
Global Market ◽  
Whole Body ◽  
Body Parts ◽  
Northern Spain ◽  
Endangered Fish ◽  
Hippocampus Guttulatus ◽  
In Captivity ◽  
Southern Portugal

The global market of dried seahorses mainly supplies Traditional Chinese Medicine and still relies on blurry trade chains that often cover less sustainable practices targeting these pricey and endangered fish. As such, reliable tools that allow the enforcement of traceability, namely to confirm the geographic origin of traded seahorses, are urgently needed. The present study evaluated the use of elemental fingerprints (EF) in the bony structures of long-snouted seahorses Hippocampus guttulatus raised in captivity in two different locations (southern Portugal and Northern Spain) to discriminate their geographic origin. The EF of different body parts of H. guttulatus were also evaluated as potential proxies for the EF of the whole body, in order to allow the analysis of damaged specimens and avoid the use of whole specimens for analysis. The contrasting EF of H. guttulatus raised in the two locations allowed their reliable discrimination. Although no single body part exactly mimicked the EF of the whole body, seahorse trunks, as well as damaged specimens, could still be correctly allocated to their geographic origin. This promising forensic approach to discriminate the geographic origin of seahorses raised in captivity should now be validated for wild conspecifics originating from different locations, as well as for other species within genus Hippocampus.

scholarly journals Statistical Analysis ofHie(Cold Sensation) andHiesho(Cold Disorder) in Kampo Clinic

2013 ◽  
Vol 2013 ◽  
pp. 1-8 ◽  
Author(s):  
Tetsuhiro Yoshino ◽  
Kotoe Katayama ◽  
Kaori Munakata ◽  
Yuko Horiba ◽  
Rui Yamaguchi ◽  
...  
Keyword(s):  
Body Part ◽  
University Hospital ◽  
Upper Body ◽  
Whole Body ◽  
Hot Flushes ◽  
Cold Sensation ◽  
Treatment Target ◽  
Lower Back ◽  
Analogue Scales ◽  
Body Heat

A cold sensation (hie) is common in Japanese women and is an important treatment target in Kampo medicine. Physicians diagnose patients as havinghiesho(cold disorder) whenhiedisturbs their daily activity. However, differences betweenhieandhieshoin men and women are not well described.Hiecan be of three types depending on body part where patients feelhie. We aimed to clarify the characteristics of patients withhieandhieshoby analyzing data from new patients seen at the Kampo Clinic at Keio University Hospital between 2008 and 2013. We collected information about patients’ subjective symptoms and their severity using visual analogue scales. Of 4,016 new patients, 2,344 complained abouthieand 524 of those were diagnosed withhiesho.Hiewas most common in legs/feet and combined with hands or lower back, rather than the whole body. Almost 30% of patients withhiefelt upper body heat symptoms like hot flushes. Cold sensation was stronger inhieshothannon-hieshopatients. Patients withhiehad more complaints. Men withhieshohad the same distribution ofhieand had symptoms similar to women. The results of our study may increase awareness ofhieshoand help doctors treathieand other symptoms.

Sign in / Sign up

Export Citation Format

Share Document