Hybrid Pooling Fusion in the BoW Pipeline

Marc Teva Law 1 Nicolas Thome 1 Matthieu Cord 1
1 MALIRE - Machine Learning and Information Retrieval
LIP6 - Laboratoire d'Informatique de Paris 6
Abstract : In the context of object and scene recognition, state-of-the-art performances are obtained with Bag of Words (BoW) models of mid-level representations computed from dense sampled local descriptors (e.g. SIFT). Several methods to combine low-level features and to set mid-level parameters have been evaluated recently for image classification.In this paper, we further investigate the impact of the main parameters in the BoW pipeline. We show that an adequate combination of several low (sampling rate, multiscale) and mid level (codebook size, normalization) parameters is decisive to reach good performances. Based on this analysis, we propose a merging scheme exploiting the specificities of edge-based descriptors. Low and high-contrast regions are pooled separately and combined to provide a powerful representation of images. Sucessful experiments are provided on the Caltech-101 and Scene-15 datasets.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01282471
Contributor : Lip6 Publications <>
Submitted on : Thursday, March 3, 2016 - 4:50:36 PM
Last modification on : Thursday, March 21, 2019 - 1:06:32 PM

Identifiers

Citation

Marc Teva Law, Nicolas Thome, Matthieu Cord. Hybrid Pooling Fusion in the BoW Pipeline. ECCV 2012 Workshop on Information fusion in Computer Vision for Concept Recognition (ECCV-IFCVCR 2012), Oct 2012, Florence, Italy. pp.355-364, ⟨10.1007/978-3-642-33885-4_36⟩. ⟨hal-01282471⟩

Share

Metrics

Record views

85