Fine-grained recognition without part annotations
- 1 June 2015
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5546-5555
- https://doi.org/10.1109/cvpr.2015.7299194
Abstract
Scaling up fine-grained recognition to all domains of fine-grained objects is a challenge the computer vision community will need to face in order to realize its goal of recognizing all object categories. Current state-of-the-art techniques rely heavily upon the use of keypoint or part annotations, but scaling up to hundreds or thousands of domains renders this annotation cost-prohibitive for all but the most important categories. In this work we propose a method for fine-grained recognition that uses no part annotations. Our method is based on generating parts using co-segmentation and alignment, which we combine in a discriminative mixture. Experimental results show its efficacy, demonstrating state-of-the-art results even when compared to methods that use part annotations during training.Keywords
This publication has 31 references indexed in Scilit:
- Rich Feature Hierarchies for Accurate Object Detection and Semantic SegmentationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2014
- Nonparametric Part Transfer for Fine-Grained RecognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2014
- ImageNet Auto-Annotation with Segmentation PropagationInternational Journal of Computer Vision, 2014
- 3D Object Representations for Fine-Grained CategorizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Unsupervised Joint Object Discovery and Segmentation in Internet ImagesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Image segmentation for large-scale subcategory flower recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Image segmentation with a bounding box priorPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- A Visual Vocabulary for Flower ClassificationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2006
- "GrabCut"ACM Transactions on Graphics, 2004
- Shape matching and object recognition using shape contextsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002