Abstract | ||
---|---|---|
Generalized zero-shot learning (GZSL) aims to recognize any test instance coming either from a known class or from a novel class that has no training instance. To synthesize training instances for novel classes and thus resolving GZSL as a common classification problem, we propose a Dissimilarity Representation Learning (DSS) method. Dissimilarity representation is to represent a specific instance in terms of its (dis)similarity to other instances in a visual or attribute based feature space. In the dissimilarity space, instances of the novel classes are synthesized by an end-to-end optimized neural network. The neural network realizes two-level feature mappings and domain adaptions in the dissimilarity space and the attribute based feature space. Experimental results on five benchmark datasets, i.e., AWA, AWA$_2$, SUN, CUB, and aPY, show that the proposed method improves the state-of-the-art with a large margin, approximately 10% gain in terms of the harmonic mean of the top-1 accuracy. Consequently, this paper establishes a new baseline for GZSL.
|
Year | DOI | Venue |
---|---|---|
2018 | 10.1145/3240508.3240686 | MM '18: ACM Multimedia Conference
Seoul
Republic of Korea
October, 2018 |
Keywords | Field | DocType |
Dissimilarity Representation, Generalized Zero-Shot Learning, Feature Mapping | Computer vision,Feature vector,Feature mapping,Pattern recognition,Computer science,Harmonic mean,Artificial intelligence,Artificial neural network,Feature learning | Conference |
ISBN | Citations | PageRank |
978-1-4503-5665-7 | 1 | 0.35 |
References | Authors | |
22 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gang Yang | 1 | 53 | 15.64 |
Jinlu Liu | 2 | 2 | 0.70 |
Jieping Xu | 3 | 4 | 1.77 |
Xirong Li | 4 | 1191 | 68.62 |