Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

FEATURE SET CONSOLIDATION FOR OBJECT REPRESENTATION BY PARTS

Abstract : Image data is growing by leaps and bounds. Machine learning based applications that run on image datasets increasingly use local image feature descriptors. In a sense, we can now visualize images as objects with features as parts. Typically there are thousands of local features per image, resulting in an explosion of feature set size for already humungous image datasets. In this paper we present a feature set consolidation strategy based on two aspects: pruning of non-discriminatory features across different object types and association of matching features for the same object type. We showcase the effectiveness of our consolidation strategy by performing classification on a building dataset. Our method reduces storage space footprint (~5%) and classification runtime (~4%), and increases classification accuracy (~2%).
Document type :
Preprints, Working Papers, ...
Complete list of metadata

Cited literature [12 references]  Display  Hide  Download

https://hal.inria.fr/hal-01708549
Contributor : Piyush Yadav <>
Submitted on : Tuesday, February 13, 2018 - 5:33:39 PM
Last modification on : Thursday, February 22, 2018 - 1:09:42 AM
Long-term archiving on: : Saturday, May 5, 2018 - 10:41:56 PM

File

FEATURE SET CONSOLIDATION FOR ...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01708549, version 1

Collections

Citation

Shamsuddin Ladha, Piyush Yadav, Shailesh Deshpande. FEATURE SET CONSOLIDATION FOR OBJECT REPRESENTATION BY PARTS. 2018. ⟨hal-01708549⟩

Share

Metrics

Record views

42

Files downloads

104