Sökning: id:"swepub:oai:DiVA.org:his-19533" >
Pedestrian’s Intent...
Pedestrian’s Intention Recognition, Fusion of Handcrafted Features in a Deep Learning Approach
-
- Hamed, Omar (författare)
- Högskolan i Skövde,Institutionen för informationsteknologi,Forskningsmiljön Informationsteknologi,Skövde Artificial Intelligence Lab (SAIL)
-
- Steinhauer, H. Joe (författare)
- Högskolan i Skövde,Institutionen för informationsteknologi,Forskningsmiljön Informationsteknologi,Skövde Artificial Intelligence Lab (SAIL)
-
(creator_code:org_t)
- Palo Alto : AAAI Press, 2021
- 2021
- Engelska.
-
Ingår i: AAAI-21 / IAAI-21 / EAAI-21 Proceedings: A virtual conference February 2-9, 2021. - Palo Alto : AAAI Press. - 9781577358664 ; , s. 15795-15796
- Relaterad länk:
-
https://ojs.aaai.org...
-
visa fler...
-
https://urn.kb.se/re...
-
visa färre...
Abstract
Ämnesord
Stäng
- The safety of vulnerable road users (VRU) is a major concernfor both advanced driver assistance systems (ADAS) and autonomousvehicle manufacturers. To guarantee people safetyon roads, autonomous vehicles must be able to detect thepresence of pedestrians, track them, and predict their intentionto cross the road. Most of the earlier work on pedestrianintention recognition focused on using either handcraftedfeatures or an end-to-end deep learning approach. In thisproject, we investigate the impact of fusing handcrafted featureswith auto learned features by using a two-stream neuralnetwork architecture. Our results show that the combined approachimproves the performance. Furthermore, the proposedmethod achieved very good results on the JAAD dataset. Dependingon whether we considered the immediate frames beforethe crossing or only half a second before that point, wereceived prediction accuracy of 91%, and 84%, respectively.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
Nyckelord
- Skövde Artificial Intelligence Lab (SAIL)
- Skövde Artificial Intelligence Lab (SAIL)
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)
Hitta via bibliotek
Till lärosätets databas