Online Monitoring for Neural Network Based Monocular Pedestrian Pose Estimation
| العنوان: | Online Monitoring for Neural Network Based Monocular Pedestrian Pose Estimation |
|---|---|
| المؤلفون: | Arjun Gupta, Luca Carlone |
| المساهمون: | Massachusetts Institute of Technology. Laboratory for Information and Decision Systems, Massachusetts Institute of Technology. Department of Aeronautics and Astronautics |
| المصدر: | arXiv ITSC |
| بيانات النشر: | IEEE, 2020. |
| سنة النشر: | 2020 |
| مصطلحات موضوعية: | FOS: Computer and information sciences, Computer Science - Machine Learning, Correctness, Computer science, media_common.quotation_subject, Computer Vision and Pattern Recognition (cs.CV), Feature extraction, Computer Science - Computer Vision and Pattern Recognition, 02 engineering and technology, computer.software_genre, Machine Learning (cs.LG), Atom (programming language), Robustness (computer science), 0202 electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering, Quality (business), Pose, media_common, Artificial neural network, business.industry, Deep learning, Image and Video Processing (eess.IV), 020207 software engineering, Electrical Engineering and Systems Science - Image and Video Processing, 020201 artificial intelligence & image processing, Artificial intelligence, Data mining, business, computer |
| الوصف: | Several autonomy pipelines now have core components that rely on deep learning approaches. While these approaches work well in nominal conditions, they tend to have unexpected and severe failure modes that create concerns when used in safety-critical applications, including self-driving cars. There are several works that aim to characterize the robustness of networks offline, but currently there is a lack of tools to monitor the correctness of network outputs online during operation. We investigate the problem of online output monitoring for neural networks that estimate 3D human shapes and poses from images. Our first contribution is to present and evaluate model-based and learning-based monitors for a human-pose-and-shape reconstruction network, and assess their ability to predict the output loss for a given test input. As a second contribution, we introduce an Adversarially-Trained Online Monitor ( ATOM ) that learns how to effectively predict losses from data. ATOM dominates model-based baselines and can detect bad outputs, leading to substantial improvements in human pose output quality. Our final contribution is an extensive experimental evaluation that shows that discarding outputs flagged as incorrect by ATOM improves the average error by 12.5%, and the worst-case error by 126.5%. Accepted to ITSC 2020 |
| وصف الملف: | application/pdf |
| اللغة: | English |
| URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_dedup___::59d784ddc61f74abc6f7b58faf2491ff https://hdl.handle.net/1721.1/137288 |
| Rights: | OPEN |
| رقم الانضمام: | edsair.doi.dedup.....59d784ddc61f74abc6f7b58faf2491ff |
| قاعدة البيانات: | OpenAIRE |
| الوصف غير متاح. |