Human keypoint detection for close proximity human-robot interaction
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F22%3A00362006" target="_blank" >RIV/68407700:21230/22:00362006 - isvavai.cz</a>
Výsledek na webu
<a href="https://doi.org/10.1109/Humanoids53995.2022.10000133" target="_blank" >https://doi.org/10.1109/Humanoids53995.2022.10000133</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/Humanoids53995.2022.10000133" target="_blank" >10.1109/Humanoids53995.2022.10000133</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Human keypoint detection for close proximity human-robot interaction
Popis výsledku v původním jazyce
We study the performance of state-of-the-art human keypoint detectors in the context of close proximity human-robot interaction. The detection in this scenario is specific in that only a subset of body parts such as hands and torso are in the field of view. In particular, (i) we survey existing datasets with human pose annotation from the perspective of close proximity images and prepare and make publicly available a new Human in Close Proximity (HiCP) dataset; (ii) we quantitatively and qualitatively compare state-of-the-art human whole-body 2D keypoint detection methods (OpenPose, MMPose, AlphaPose, Detectron2) on this dataset; (iii) since accurate detection of hands and fingers is critical in applications with handovers, we evaluate the performance of the MediaPipe hand detector; (iv) we deploy the algorithms on a humanoid robot with an RGB-D camera on its head and evaluate the performance in 3D human keypoint detection. A motion capture system is used as reference. The best performing whole-body keypoint detectors in close proximity were MMPose and AlphaPose, but both had difficulty with finger detection. Thus, we propose a combination of MMPose or AlphaPose for the body and MediaPipe for the hands in a single framework providing the most accurate and robust detection. We also analyse the failure modes of individual detectors---for example, to what extent the absence of the head of the person in the image degrades performance. Finally, we demonstrate the framework in a scenario where a humanoid robot interacting with a person uses the detected 3D keypoints for whole-body avoidance maneuvers.
Název v anglickém jazyce
Human keypoint detection for close proximity human-robot interaction
Popis výsledku anglicky
We study the performance of state-of-the-art human keypoint detectors in the context of close proximity human-robot interaction. The detection in this scenario is specific in that only a subset of body parts such as hands and torso are in the field of view. In particular, (i) we survey existing datasets with human pose annotation from the perspective of close proximity images and prepare and make publicly available a new Human in Close Proximity (HiCP) dataset; (ii) we quantitatively and qualitatively compare state-of-the-art human whole-body 2D keypoint detection methods (OpenPose, MMPose, AlphaPose, Detectron2) on this dataset; (iii) since accurate detection of hands and fingers is critical in applications with handovers, we evaluate the performance of the MediaPipe hand detector; (iv) we deploy the algorithms on a humanoid robot with an RGB-D camera on its head and evaluate the performance in 3D human keypoint detection. A motion capture system is used as reference. The best performing whole-body keypoint detectors in close proximity were MMPose and AlphaPose, but both had difficulty with finger detection. Thus, we propose a combination of MMPose or AlphaPose for the body and MediaPipe for the hands in a single framework providing the most accurate and robust detection. We also analyse the failure modes of individual detectors---for example, to what extent the absence of the head of the person in the image degrades performance. Finally, we demonstrate the framework in a scenario where a humanoid robot interacting with a person uses the detected 3D keypoints for whole-body avoidance maneuvers.
Klasifikace
Druh
D - Stať ve sborníku
CEP obor
—
OECD FORD obor
20204 - Robotics and automatic control
Návaznosti výsledku
Projekt
<a href="/cs/project/EF16_019%2F0000765" target="_blank" >EF16_019/0000765: Výzkumné centrum informatiky</a><br>
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach
Ostatní
Rok uplatnění
2022
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Údaje specifické pro druh výsledku
Název statě ve sborníku
2022 IEEE-RAS International Conference on Humanoid Robots (Humanoids)
ISBN
979-8-3503-0979-9
ISSN
2164-0572
e-ISSN
2164-0580
Počet stran výsledku
8
Strana od-do
450-457
Název nakladatele
IEEE
Místo vydání
Piscataway, NJ
Místo konání akce
Okinawa
Datum konání akce
28. 11. 2022
Typ akce podle státní příslušnosti
WRD - Celosvětová akce
Kód UT WoS článku
000925894300059