TY - GEN
T1 - Reliable attribute-based object recognition using high predictive value classifiers
AU - Luan, Wentao
AU - Yang, Yezhou
AU - Fermüller, Cornelia
AU - Baras, John S.
N1 - Funding Information:
This work was funded by the support of DARPA (through ARO) grant W911NF1410384, by NSF through grants CNS-1544787 and SMA-1540917 and Samsung under the GRO program (N020477, 355022).
PY - 2016
Y1 - 2016
N2 - We consider the problem of object recognition in 3D using an ensemble of attribute-based classifiers. We propose two new concepts to improve classification in practical situations, and show their implementation in an approach implemented for recognition from point-cloud data. First, the viewing conditions can have a strong influence on classification performance. We study the impact of the distance between the camera and the object and propose an approach to fusing multiple attribute classifiers, which incorporates distance into the decision making. Second, lack of representative training samples often makes it difficult to learn the optimal threshold value for best positive and negative detection rate. We address this issue, by setting in our attribute classifiers instead of just one threshold value, two threshold values to distinguish a positive, a negative and an uncertainty class, and we prove the theoretical correctness of this approach. Empirical studies demonstrate the effectiveness and feasibility of the proposed concepts.
AB - We consider the problem of object recognition in 3D using an ensemble of attribute-based classifiers. We propose two new concepts to improve classification in practical situations, and show their implementation in an approach implemented for recognition from point-cloud data. First, the viewing conditions can have a strong influence on classification performance. We study the impact of the distance between the camera and the object and propose an approach to fusing multiple attribute classifiers, which incorporates distance into the decision making. Second, lack of representative training samples often makes it difficult to learn the optimal threshold value for best positive and negative detection rate. We address this issue, by setting in our attribute classifiers instead of just one threshold value, two threshold values to distinguish a positive, a negative and an uncertainty class, and we prove the theoretical correctness of this approach. Empirical studies demonstrate the effectiveness and feasibility of the proposed concepts.
UR - http://www.scopus.com/inward/record.url?scp=84990041919&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84990041919&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-46487-9_49
DO - 10.1007/978-3-319-46487-9_49
M3 - Conference contribution
AN - SCOPUS:84990041919
SN - 9783319464862
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 801
EP - 815
BT - Computer Vision - 14th European Conference, ECCV 2016, Proceedings
A2 - Leibe, Bastian
A2 - Matas, Jiri
A2 - Sebe, Nicu
A2 - Welling, Max
PB - Springer Verlag
T2 - 14th European Conference on Computer Vision, ECCV 2016
Y2 - 8 October 2016 through 16 October 2016
ER -