Segmentation-based multi-class semantic object detection

Abstract

In this paper we study the problem of the detection of semantic objects from known categories in images. Unlike existing techniques which operate at the pixel or at a patch level for recognition, we propose to rely on the categorization of image segments. Recent work has highlighted that image segments provide a sound support for visual object class recognition. In this work, we use image segments as primitives to extract robust features and train detection models for a predefined set of categories. Several segmentation algorithms are benchmarked and their performances for segment recognition are compared. We then propose two methods for enhancing the segments classification, one based on the fusion of the classification results obtained with the different segmentations, the other one based on the optimization of the global labelling by correcting local ambiguities between neighbor segments. We use as a benchmark the Microsoft MSRC-21 image database and show that our method competes with the current state-of-the-art.

Improving Semantic Segmentation via Video Propagation and Label Relaxation

“…One is the potential error propagation from edge estimation and the other is fitting extremely hard boundary cases may lead to over-fitting at the test stage. There is also literature focusing on structure modeling to obtain better boundary localization, such as affinity field [21], random walk [5], relaxation labelling [37], boundary neural fields [4], etc. However, none of these methods deals directly with boundary pixels but they instead attempt to model the interactions between segments along object boundaries.…”

A Review: Segmentation Based Objects Analysis in RGB Colored Image using Matlab Simulation

“…They propose two methods for enhancing the segments classification, one based on the fusion of the classification results obtained with the different segmentations, the other one based on the optimization of the global labelling by way of correcting local ambiguities between neighbour segments. They use as a benchmark the Microsoft MSRC-21 image database and display that approach competes with the current state-of-the-art [2].…”

Optimizations on semantic environment management: An application for humanoid robot home assistance

“…That way, a semantic-based recognition offers a more human alike approach. Despite of this, introduced solutions requires a semantic classification of the sensor information [23] [5]. As a result, semantic map definitions encourages human interaction and promotes its application on service tasks and housework [18].…”