TY - GEN
T1 - Learning from Synthetic Vehicles
AU - Kim, Tae Soo
AU - Shim, Bohoon
AU - Peven, Michael
AU - Qiu, Weichao
AU - Yuille, Alan
AU - Hager, Gregory D.
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - In this paper, we release the Simulated Articulated VEhicles Dataset (SAVED) which contains images of synthetic vehicles with moveable vehicle parts. SAVED consists of images that are much more relevant for vehicle-related pattern-recognition tasks than other popular pretraining datasets such as ImageNet. Compared to a model initialized with ImageNet weights, we show that a model pretrained using SAVED leads to much better performance when recognizing vehicle parts and orientation directly from an image. We also find that a multi-task pretraining approach using fine-grained geometric signals available in SAVED leads to significant improvements in performance. By pretraining on SAVED instead of ImageNet, we reduce the error rate of one of the state of the art vehicle orientation estimators by 51.2% when tested on real images. We release SAVED and instructions on its usage here 1.1https://taesoo-kim.github.io/
AB - In this paper, we release the Simulated Articulated VEhicles Dataset (SAVED) which contains images of synthetic vehicles with moveable vehicle parts. SAVED consists of images that are much more relevant for vehicle-related pattern-recognition tasks than other popular pretraining datasets such as ImageNet. Compared to a model initialized with ImageNet weights, we show that a model pretrained using SAVED leads to much better performance when recognizing vehicle parts and orientation directly from an image. We also find that a multi-task pretraining approach using fine-grained geometric signals available in SAVED leads to significant improvements in performance. By pretraining on SAVED instead of ImageNet, we reduce the error rate of one of the state of the art vehicle orientation estimators by 51.2% when tested on real images. We release SAVED and instructions on its usage here 1.1https://taesoo-kim.github.io/
UR - http://www.scopus.com/inward/record.url?scp=85126773378&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85126773378&partnerID=8YFLogxK
U2 - 10.1109/WACVW54805.2022.00056
DO - 10.1109/WACVW54805.2022.00056
M3 - Conference contribution
AN - SCOPUS:85126773378
T3 - Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022
SP - 500
EP - 508
BT - Proceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops, WACVW 2022
Y2 - 4 January 2022 through 8 January 2022
ER -