TY  - JOUR
PY  - 2023//
TI  - Detecting road obstacles by erasing them
JO  - IEEE transactions on pattern analysis and machine intelligence
A1  - Lis, Krzysztof
A1  - Honari, Sina
A1  - Fua, Pascal
A1  - Salzmann, Mathieu
SP  - ePub
EP  - ePub
VL  - ePub
IS  - ePub
N2  - Vehicles can encounter a myriad of obstacles on the road, and it is impossible to record them all beforehand to train a detector. Instead, we select image patches and inpaint them with the surrounding road texture, which tends to remove obstacles from those patches. We then use a network trained to recognize discrepancies between the original patch and the inpainted one, which signals an erased obstacle.<p />  <p>Language: en</p>
LA  - en
SN  - 0162-8828
UR  - http://dx.doi.org/10.1109/TPAMI.2023.3335152
ID  - ref1
ER  -