Keypoint Density-based Region Proposal for Fine-Grained Object Detection and Classification using Regions with Convolutional Neural Network Features.
J. T. TurnerKalyan Moy GuptaBrendan MorrisDavid W. AhaPublished in: CoRR (2016)
Keyphrases
- fine grained
- convolutional neural network
- face detection
- object detection
- textural features
- coarse grained
- image features
- feature vectors
- keypoints
- classification accuracy
- feature set
- feature extraction
- feature space
- image regions
- classification method
- access control
- object region
- tightly coupled
- image classification
- text classification
- data lineage
- adjacent regions
- arbitrary shape
- salient regions
- input image
- sift descriptors
- region of interest
- texture features
- object recognition
- gray level co occurrence matrix
- object categories
- pairwise
- segmented regions
- region features
- feature selection