Segmentation-based multi-class semantic object detection

Vieux, Remi; Benois-Pineau, Jenny; Domenger, Jean‐Philippe; Braquelaire, Achille

doi:10.1007/s11042-010-0611-2

Cited by 25 publications

(9 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…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.…”

Section: Related Workmentioning

confidence: 99%

Improving Semantic Segmentation via Video Propagation and Label Relaxation

Zhu

Sapra

Reda

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

393

309

View full text Add to dashboard Cite

Semantic segmentation requires large amounts of pixelwise annotations to learn accurate models. In this paper, we present a video prediction-based methodology to scale up training sets by synthesizing new training samples in order to improve the accuracy of semantic segmentation networks. We exploit video prediction models' ability to predict future frames in order to also predict future labels. A joint propagation strategy is also proposed to alleviate mis-alignments in synthesized samples. We demonstrate that training segmentation models on datasets augmented by the synthesized samples leads to significant improvements in accuracy. Furthermore, we introduce a novel boundary label relaxation technique that makes training robust to annotation noise and propagation artifacts along object boundaries. Our proposed methods achieve state-of-the-art mIoUs of 83.5% on Cityscapes and 82.9% on CamVid. Our single model, without model ensembles, achieves 72.8% mIoU on the KITTI semantic segmentation test set, which surpasses the winning entry of the ROB challenge 2018. Our code and videos can be found at https://nv-adlr.github. io/publication/2018-Segmentation.

show abstract

Section: Related Workmentioning

confidence: 99%

Improving Semantic Segmentation via Video Propagation and Label Relaxation

Zhu

Sapra

Reda

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

393

309

View full text Add to dashboard Cite

show abstract

“…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].…”

Section: Literature Reviewmentioning

confidence: 99%

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

Diksha¹,

Chhabra²

2019

International Journal of Advanced Research in Computer and Comm

View full text Add to dashboard Cite

Object Analysis is one of the great challenges of computer vision. The RGB color models are an additive coloration version in wherein red, green and blue light are added collectively in various way to reproduce a large array of colors. The primary goal of the RGB color model is for sensing, representation and show image in electronics systems like televisions and computers, though it's been used in conventional images. There are general phases that define the flaws the working on these Techniques. The proposed work has been explained the concept of the Image Segmentation means to analysis object in RGB colored and retrieve the edge of the image through image segmentation. The analysis of the different objects and quality of image can be implement using image segmentation process.

show abstract

“…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].…”

Section: Environment Description Techniquesmentioning

confidence: 99%

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

Munera

Posadas-Yagüe

Poza-Lujan

et al. 2014

2014 IEEE-RAS International Conference on Humanoid Robots

View full text Add to dashboard Cite

Abstract-This article introduces some optimization mechanisms focused on environment management, object recognition, and environment interaction. Although the generality of the presented system, this work will be focused on its application on home assistance humanoid robots. For this purpose, a generic environment formalization procedure for semantic scenery description is introduced. As the main contribution of this work, some techniques for a more efficient use of the environment knowledge are proposed. That way, the application of an areabased discrimination mechanism will avoid to process large amounts of data, useless in the current context, improving the object recognition, and characterizing the available interactions in the current area. Finally, the formalized description, and the optimization procedure, will be tested and verified on a specific home scenario using a humanoid robot.

show abstract

Segmentation-based multi-class semantic object detection

Cited by 25 publications

References 30 publications

Improving Semantic Segmentation via Video Propagation and Label Relaxation

Improving Semantic Segmentation via Video Propagation and Label Relaxation

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

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

Contact Info

Product

Resources

About