@techreport{
 title = {VRU Pose-SSD:  Multiperson  Pose  Estimation  For  Automated  Driving},
 type = {techreport},
 year = {2021},
 keywords = {EMRG: Autonomous Vehicles,EMRG: Computer Vision,EMRG: Pose Estimation},
 websites = {www.aaai.org},
 id = {fb42722c-533f-369d-a526-5a1b7c8d96ca},
 created = {2024-11-06T15:57:17.999Z},
 file_attached = {true},
 profile_id = {f1f70cad-e32d-3de2-a3c0-be1736cb88be},
 group_id = {5ec9cc91-a5d6-3de5-82f3-3ef3d98a89c1},
 last_modified = {2024-11-07T07:25:41.059Z},
 read = {true},
 starred = {false},
 authored = {false},
 confirmed = {false},
 hidden = {false},
 folder_uuids = {4cda9297-f98e-4246-88d6-ffeeade205c3},
 private_publication = {false},
 abstract = {We present a fast and efficient approach for joint person detection and pose estimation optimized for automated driving (AD) in urban scenarios. We use a multitask weight sharing architecture to jointly train detection and pose estimation. This modular architecture allows us to accommodate different downstream tasks in the future. By systematic large-scale experiments on the Tsinghua-Daimler Urban Pose Dataset (TDUP), we obtain multiple models with varying accuracy-speed trade-offs. We then quantize and optimize our network for deployment and present a detailed analysis of the efficacy of the algorithm. We introduce a two-stage evaluation strategy, which is more suitable for AD and achieves a significant performance improvement in comparison to state-of-the-art approaches. Our optimized model runs at 52 fps on full HD images and still reaches a competitive performance of 32.25 LAMR. We are confident that our work serves as an enabler to tackle higher-level tasks like VRU intention estimation and gesture recognition, which rely on stable pose estimates and will play a crucial role in future AD systems.},
 bibtype = {techreport},
 author = {Kumar, Chandan and Ramesh, Jayanth and Chakraborty, Bodhisattwa and Raman, Renjith and Weinrich, Christoph and Mundhada, Anurag and Jain, Arjun and Flohr, Fabian B}
}

{"_id":"DL7RWtwEnwstXT4nx","bibbaseid":"kumar-ramesh-chakraborty-raman-weinrich-mundhada-jain-flohr-vruposessdmultipersonposeestimationforautomateddriving-2021","author_short":["Kumar, C.","Ramesh, J.","Chakraborty, B.","Raman, R.","Weinrich, C.","Mundhada, A.","Jain, A.","Flohr, F., B."],"bibdata":{"title":"VRU Pose-SSD: Multiperson Pose Estimation For Automated Driving","type":"techreport","year":"2021","keywords":"EMRG: Autonomous Vehicles,EMRG: Computer Vision,EMRG: Pose Estimation","websites":"www.aaai.org","id":"fb42722c-533f-369d-a526-5a1b7c8d96ca","created":"2024-11-06T15:57:17.999Z","file_attached":"true","profile_id":"f1f70cad-e32d-3de2-a3c0-be1736cb88be","group_id":"5ec9cc91-a5d6-3de5-82f3-3ef3d98a89c1","last_modified":"2024-11-07T07:25:41.059Z","read":"true","starred":false,"authored":false,"confirmed":false,"hidden":false,"folder_uuids":"4cda9297-f98e-4246-88d6-ffeeade205c3","private_publication":false,"abstract":"We present a fast and efficient approach for joint person detection and pose estimation optimized for automated driving (AD) in urban scenarios. We use a multitask weight sharing architecture to jointly train detection and pose estimation. This modular architecture allows us to accommodate different downstream tasks in the future. By systematic large-scale experiments on the Tsinghua-Daimler Urban Pose Dataset (TDUP), we obtain multiple models with varying accuracy-speed trade-offs. We then quantize and optimize our network for deployment and present a detailed analysis of the efficacy of the algorithm. We introduce a two-stage evaluation strategy, which is more suitable for AD and achieves a significant performance improvement in comparison to state-of-the-art approaches. Our optimized model runs at 52 fps on full HD images and still reaches a competitive performance of 32.25 LAMR. We are confident that our work serves as an enabler to tackle higher-level tasks like VRU intention estimation and gesture recognition, which rely on stable pose estimates and will play a crucial role in future AD systems.","bibtype":"techreport","author":"Kumar, Chandan and Ramesh, Jayanth and Chakraborty, Bodhisattwa and Raman, Renjith and Weinrich, Christoph and Mundhada, Anurag and Jain, Arjun and Flohr, Fabian B","bibtex":"@techreport{\n title = {VRU Pose-SSD: Multiperson Pose Estimation For Automated Driving},\n type = {techreport},\n year = {2021},\n keywords = {EMRG: Autonomous Vehicles,EMRG: Computer Vision,EMRG: Pose Estimation},\n websites = {www.aaai.org},\n id = {fb42722c-533f-369d-a526-5a1b7c8d96ca},\n created = {2024-11-06T15:57:17.999Z},\n file_attached = {true},\n profile_id = {f1f70cad-e32d-3de2-a3c0-be1736cb88be},\n group_id = {5ec9cc91-a5d6-3de5-82f3-3ef3d98a89c1},\n last_modified = {2024-11-07T07:25:41.059Z},\n read = {true},\n starred = {false},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n folder_uuids = {4cda9297-f98e-4246-88d6-ffeeade205c3},\n private_publication = {false},\n abstract = {We present a fast and efficient approach for joint person detection and pose estimation optimized for automated driving (AD) in urban scenarios. We use a multitask weight sharing architecture to jointly train detection and pose estimation. This modular architecture allows us to accommodate different downstream tasks in the future. By systematic large-scale experiments on the Tsinghua-Daimler Urban Pose Dataset (TDUP), we obtain multiple models with varying accuracy-speed trade-offs. We then quantize and optimize our network for deployment and present a detailed analysis of the efficacy of the algorithm. We introduce a two-stage evaluation strategy, which is more suitable for AD and achieves a significant performance improvement in comparison to state-of-the-art approaches. Our optimized model runs at 52 fps on full HD images and still reaches a competitive performance of 32.25 LAMR. We are confident that our work serves as an enabler to tackle higher-level tasks like VRU intention estimation and gesture recognition, which rely on stable pose estimates and will play a crucial role in future AD systems.},\n bibtype = {techreport},\n author = {Kumar, Chandan and Ramesh, Jayanth and Chakraborty, Bodhisattwa and Raman, Renjith and Weinrich, Christoph and Mundhada, Anurag and Jain, Arjun and Flohr, Fabian B}\n}","author_short":["Kumar, C.","Ramesh, J.","Chakraborty, B.","Raman, R.","Weinrich, C.","Mundhada, A.","Jain, A.","Flohr, F., B."],"urls":{"Paper":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c/file/1adadc85-264a-acd5-3e0e-4db0d5afbdbe/17800_Article_Text_21294_1_2_20210518.pdf.pdf","Website":"www.aaai.org"},"biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","bibbaseid":"kumar-ramesh-chakraborty-raman-weinrich-mundhada-jain-flohr-vruposessdmultipersonposeestimationforautomateddriving-2021","role":"author","keyword":["EMRG: Autonomous Vehicles","EMRG: Computer Vision","EMRG: Pose Estimation"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"techreport","biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","dataSources":["2252seNhipfTmjEBQ"],"keywords":["emrg: autonomous vehicles","emrg: computer vision","emrg: pose estimation"],"search_terms":["vru","pose","ssd","multiperson","pose","estimation","automated","driving","kumar","ramesh","chakraborty","raman","weinrich","mundhada","jain","flohr"],"title":"VRU Pose-SSD: Multiperson Pose Estimation For Automated Driving","year":2021}