Region proposal
A region proposal is a candidate image region identified as potentially containing an object for subsequent classification and localization.
Math and Architectures of Deep Learning
A candidate image region proposed as possibly containing an object for later classification and localization.
- How it is used
- Narrows object detection into candidate-region generation followed by classification and localization.
- How deep it goes
- Explains proposal generation, shared feature maps, RoI pooling, and two training strategies, while deferring selective-search details.
- Examples
- Uses roughly 2,000 selective-search proposals and maps a 256 × 256 RoI to a 16 × 16 feature-map region.
- What you can do after
- Evaluate/understand how candidate regions become class scores and refined bounding boxes in a two-stage detector.
Read how Math and Architectures of Deep Learning explains region proposal →
Inside Deep Learning: Math, Algorithms, Models
An object-detection approach that first proposes candidate regions likely to contain objects before classifying them.
- How it is used
- Uses candidate regions to produce variable-number object-detection predictions with relatively inexpensive bounding-box labels.
- How deep it goes
- Explains the architecture and transferable principles while deferring minute details of a specific implementation.
- Examples
- Uses bounding-box annotation and a built-in PyTorch detector as concrete examples.
- What you can do after
- Evaluate/understand region proposal detection and its transferable architectural principles.
Read how Inside Deep Learning: Math, Algorithms, Models explains region proposal →
Related concepts
- Region Proposal Networkproduced by · A Region Proposal Network is a neural-network module that generates candidate object regions from image features for a two-stage detector.
- Faster R-CNNhas part · Faster R-CNN is an object-detection architecture that uses a learned region proposal network and detection heads to localize and classify objects in images.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.