Pith. sign in

REVIEW

Spatial Semantic Regularisation for Large Scale Object Detection

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1510.02949 v1 pith:DSUTCF5N submitted 2015-10-10 cs.CV

Spatial Semantic Regularisation for Large Scale Object Detection

classification cs.CV
keywords detectionscalesemanticspatialclasseslargethousandsacross
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Large scale object detection with thousands of classes introduces the problem of many contradicting false positive detections, which have to be suppressed. Class-independent non-maximum suppression has traditionally been used for this step, but it does not scale well as the number of classes grows. Traditional non-maximum suppression does not consider label- and instance-level relationships nor does it allow an exploitation of the spatial layout of detection proposals. We propose a new multi-class spatial semantic regularisation method based on affinity propagation clustering, which simultaneously optimises across all categories and all proposed locations in the image, to improve both the localisation and categorisation of selected detection proposals. Constraints are shared across the labels through the semantic WordNet hierarchy. Our approach proves to be especially useful in large scale settings with thousands of classes, where spatial and semantic interactions are very frequent and only weakly supervised detectors can be built due to a lack of bounding box annotations. Detection experiments are conducted on the ImageNet and COCO dataset, and in settings with thousands of detected categories. Our method provides a significant precision improvement by reducing false positives, while simultaneously improving the recall.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.