AWhat changed
Placed fixed-bin spatial pyramid pooling above convolutional maps to produce fixed-length features from varying image/region sizes and reuse whole-image convolutions across detection proposals.
Scientific scope
Placed fixed-bin spatial pyramid pooling above convolutional maps to produce fixed-length features from varying image/region sizes and reuse whole-image convolutions across detection proposals. Spatial pyramid pooling predates this deep-network integration. Original training uses selected fixed input sizes with shared weights; acceptance of variable shapes is not proven accuracy invariance across arbitrary scales. Detector speedups are source benchmark conditions, not universal end-to-end ratios.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Spatial pyramid pooling predates this deep-network integration. Original training uses selected fixed input sizes with shared weights; acceptance of variable shapes is not proven accuracy invariance across arbitrary scales. Detector speedups are source benchmark conditions, not universal end-to-end ratios.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction