TY - JOUR
T1 - Ellipse R-CNN
T2 - Learning to Infer Elliptical Object from Clustering and Occlusion
AU - Dong, Wenbo
AU - Roy, Pravakar
AU - Peng, Cheng
AU - Isler, Volkan
N1 - Publisher Copyright:
© 1992-2012 IEEE.
PY - 2021
Y1 - 2021
N2 - Images of heavily occluded objects in cluttered scenes, such as fruit clusters in trees, are hard to segment. To further retrieve the 3D size and 6D pose of each individual object in such cases, bounding boxes are not reliable from multiple views since only a little portion of the object's geometry is captured. We introduce the first CNN-based ellipse detector, called Ellipse R-CNN, to represent and infer occluded objects as ellipses. We first propose a robust and compact ellipse regression based on the Mask R-CNN architecture for elliptical object detection. Our method can infer the parameters of multiple elliptical objects even they are occluded by other neighboring objects. For better occlusion handling, we exploit refined feature regions for the regression stage, and integrate the U-Net structure for learning different occlusion patterns to compute the final detection score. The correctness of ellipse regression is validated through experiments performed on synthetic data of clustered ellipses. We further quantitatively and qualitatively demonstrate that our approach outperforms the state-of-the-art model (i.e., Mask R-CNN followed by ellipse fitting) and its three variants on both synthetic and real datasets of occluded and clustered elliptical objects.
AB - Images of heavily occluded objects in cluttered scenes, such as fruit clusters in trees, are hard to segment. To further retrieve the 3D size and 6D pose of each individual object in such cases, bounding boxes are not reliable from multiple views since only a little portion of the object's geometry is captured. We introduce the first CNN-based ellipse detector, called Ellipse R-CNN, to represent and infer occluded objects as ellipses. We first propose a robust and compact ellipse regression based on the Mask R-CNN architecture for elliptical object detection. Our method can infer the parameters of multiple elliptical objects even they are occluded by other neighboring objects. For better occlusion handling, we exploit refined feature regions for the regression stage, and integrate the U-Net structure for learning different occlusion patterns to compute the final detection score. The correctness of ellipse regression is validated through experiments performed on synthetic data of clustered ellipses. We further quantitatively and qualitatively demonstrate that our approach outperforms the state-of-the-art model (i.e., Mask R-CNN followed by ellipse fitting) and its three variants on both synthetic and real datasets of occluded and clustered elliptical objects.
KW - 3D object localization
KW - Ellipse regression
KW - convolutional neural networks
KW - object detection
KW - occlusion handling
UR - http://www.scopus.com/inward/record.url?scp=85099727051&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85099727051&partnerID=8YFLogxK
U2 - 10.1109/TIP.2021.3050673
DO - 10.1109/TIP.2021.3050673
M3 - Article
C2 - 33471755
AN - SCOPUS:85099727051
SN - 1057-7149
VL - 30
SP - 2193
EP - 2206
JO - IEEE Transactions on Image Processing
JF - IEEE Transactions on Image Processing
M1 - 9329165
ER -