Evaluation of Output Embeddings for Fine-Grained Image Classification

Abstract

Image classification has advanced significantly in recent years with the availability of large-scale image sets. However, fine-grained classification remains a major challenge due to the annotation cost of large numbers of fine-grained categories. This project shows that compelling classification performance can be achieved on such categories even without labeled training data. Given image and class embeddings, we learn a compatibility function such that matching embeddings are assigned a higher score than mismatching ones; zero-shot classification of an image proceeds by finding the label yielding the highest joint compatibility score. We use state-of-the-art image features and focus on different supervised attributes and unsupervised output embeddings either derived from hierarchies or learned from unlabeled text corpora. We establish a substantially improved state-of-the-art on the Animals with Attributes and Caltech-UCSD Birds datasets. Most encouragingly, we demonstrate that purely unsupervised output embeddings (learned from Wikipedia and improved with fine-grained text) achieve compelling results, even outperforming the previous supervised state-of-the-art. By combining different output embeddings, we further improve results.

@inproceedings {ARWLS15, title = {Evaluation of Output Embeddings for Fine-Grained Image Classification}, booktitle = {IEEE Computer Vision and Pattern Recognition}, year = {2015}, author = {Zeynep Akata and Scott Reed and Daniel Walter and Honglak Lee and Bernt Schiele} }

References in corpus (1)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Cited by in corpus (23)

• Label-Embedding for Image Classification
• A Unified approach for Conventional Zero-shot, Generalized Zero-shot and Few-shot Learning
• AI Challenger : A Large-scale Dataset for Going Deeper in Image Understanding
• Fine-graind Image Classification via Combining Vision and Language
• Semantics-Guided Contrastive Network for Zero-Shot Object detection
• Multi-Task Zero-Shot Action Recognition with Prioritised Data Augmentation
• Context-aware Feature Generation for Zero-shot Semantic Segmentation
• Zero-Shot Detection
• Fine-Grained Object Recognition and Zero-Shot Learning in Remote Sensing Imagery
• Cluster-based Zero-shot learning for multivariate data
• Zero-Shot Learning posed as a Missing Data Problem
• ElasticTrainer: Speeding Up On-Device Training with Runtime Elastic Tensor Selection
• Disentangling Semantic-to-visual Confusion for Zero-shot Learning
• Improving deep learning with prior knowledge and cognitive models: A survey on enhancing explainability, adversarial robustness and zero-shot learning
• Vocabulary-informed Zero-shot and Open-set Learning
• Transductive Zero-Shot Hashing for Multilabel Image Retrieval
• Online Lifelong Generalized Zero-Shot Learning
• From Classical to Generalized Zero-Shot Learning: a Simple Adaptation Process
• Tell me what you see: A zero-shot action recognition method based on natural language descriptions
• An Integral Projection-based Semantic Autoencoder for Zero-Shot Learning
• Region Semantically Aligned Network for Zero-Shot Learning
• Transfer feature generating networks with semantic classes structure for zero-shot learning
• Structure propagation for zero-shot learning