Deep randomized embedding
Download as PPTX, PDF0 likes101 views
The document discusses a state-of-the-art model on the CARS196 dataset that uses proxy NCA loss to train embeddings for fine-grained vehicle classification. It trains multiple embeddings, where the dimension of each embedding equals the number of meta-classes and the number of embeddings is L. The model achieves higher performance than a 48-model ensemble by viewing samples within a meta-class as sharing a latent attribute and enforcing diversity of clusters across embeddings. It also questions whether attribute labels are truly needed to enhance feature learning.
1 of 7
Download to read offline
More Related Content
What's hot (7)
Ad
More from 哲东 郑 (20)
Ad
Recently uploaded (20)
Deep randomized embedding
- 1. ECCV2018， sota on CARS196, 9 pages
- 3. The number of embeddings: L The dimension of each embedding: D D equals the number of meta-classes Proxy NCA loss for training each embedding
- 4. The dimension of each embedding: D The number of embeddings: L
- 5. The dimension of each embedding: D The number of embeddings: L D L Higher than our SSR with 48 models ensemble
- 6. Discussion1: Do we really need attribute to enhance feature learning? Samples within a meta class can be viewed as sharing a latent attribute. So meta classes corresponds to randomized attributes
- 7. Discussion2: In hidden layers, we may expect some clusters within the dataset. A cluster may be viewed as a meta class. employing meta class = enforcing diversity of clustering? Discussion3: Encoding the original one-hot label into a sequential label. Using L-2 loss (or KLDiv loss, etc.) for learning the embedding brings about a similar improvement?