TY - GEN
T1 - Beyond spatial pooling
T2 - IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015
AU - Li, Chi
AU - Reiter, Austin
AU - Hager, Gregory D.
N1 - Publisher Copyright:
© 2015 IEEE.
Copyright:
Copyright 2017 Elsevier B.V., All rights reserved.
PY - 2015/10/14
Y1 - 2015/10/14
N2 - Object recognition systems have shown great progress over recent years. However, creating object representations that are robust to changes in viewpoint while capturing local visual details continues to be a challenge. In particular, recent convolutional architectures employ spatial pooling to achieve scale and shift invariances, but they are still sensitive to out-of-plane rotations. In this paper, we formulate a probabilistic framework for analyzing the performance of pooling. This framework suggests two directions for improvement. First, we apply multiple scales of filters coupled with different pooling granularities, and second we make use of color as an additional pooling domain, thereby reducing the sensitivity to spatial deformations. We evaluate our algorithm on the object instance recognition task using two independent publicly available RGB-D datasets, and demonstrate significant improvements over the current state-of-the-art. In addition, we present a new dataset for industrial objects to further validate the effectiveness of our approach versus other state-of-the-art approaches for object recognition using RGB-D data.
AB - Object recognition systems have shown great progress over recent years. However, creating object representations that are robust to changes in viewpoint while capturing local visual details continues to be a challenge. In particular, recent convolutional architectures employ spatial pooling to achieve scale and shift invariances, but they are still sensitive to out-of-plane rotations. In this paper, we formulate a probabilistic framework for analyzing the performance of pooling. This framework suggests two directions for improvement. First, we apply multiple scales of filters coupled with different pooling granularities, and second we make use of color as an additional pooling domain, thereby reducing the sensitivity to spatial deformations. We evaluate our algorithm on the object instance recognition task using two independent publicly available RGB-D datasets, and demonstrate significant improvements over the current state-of-the-art. In addition, we present a new dataset for industrial objects to further validate the effectiveness of our approach versus other state-of-the-art approaches for object recognition using RGB-D data.
UR - http://www.scopus.com/inward/record.url?scp=84959254410&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84959254410&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2015.7299125
DO - 10.1109/CVPR.2015.7299125
M3 - Conference contribution
AN - SCOPUS:84959254410
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 4913
EP - 4922
BT - IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015
PB - IEEE Computer Society
Y2 - 7 June 2015 through 12 June 2015
ER -