What's Wrong with That Object? Identifying Images of Unusual Objects by Modelling the Detection Score Distribution.
Peng WangLingqiao LiuChunhua ShenZi HuangAnton van den HengelHeng Tao ShenPublished in: CVPR (2016)
Keyphrases
- detecting objects
- bounding box
- multiple objects
- individual objects
- real objects
- partial occlusion
- real world objects
- cluttered images
- objects represented
- target object
- object detectors
- object detection
- visual appearance
- background clutter
- object identification
- deformable objects
- spatial relationships
- real world scenes
- object models
- partially occluded objects
- cluttered background
- image regions
- complex scenes
- complex background
- three dimensional objects
- object shape
- cluttered scenes
- data objects
- viewing angle
- quad trees
- visual objects
- recognizing objects
- three dimensional
- object localization
- object parts
- similar objects
- viewing conditions
- geometric information
- lighting conditions
- object categories
- object classes
- object model
- d objects
- keypoints
- test images
- man made objects
- object shapes
- multiple images
- input image
- moving objects
- relative position
- object features
- image features
- figure ground segmentation
- range images
- viewing direction
- object segmentation
- object class
- rigid objects
- visual input
- anomaly detection
- object representation
- spatial information
- articulated objects
- object appearance
- geometric constraints
- spatial coordinates
- image retrieval
- object recognition