UniParser: Multi-Human Parsing With Unified Correlation Representation Learning (2023-10-13T00:00:00.000000Z)

TL;DR

By unifying instance-level and category-level output, UniParser circumvents manually designed post-processing techniques and surpasses state-of-the-art methods, achieving 49.3% AP on MHPv2.0 and 60.4% AP on CIHP.

Abstract

Multi-human parsing is an image segmentation task necessitating both instance-level and fine-grained category-level information. However, prior research has typically processed these two types of information through distinct branch types and output formats, leading to inefficient and redundant frameworks. This paper introduces UniParser, which integrates instance-level and category-level representations in three key aspects: 1) we propose a unified correlation representation learning approach, allowing our network to learn instance and category features within the cosine space; 2) we unify the form of outputs of each modules as pixel-level results while supervising instance and category features using a homogeneous label accompanied by an auxiliary loss; and 3) we design a joint optimization procedure to fuse instance and category representations. By unifying instance-level and category-level output, UniParser circumvents manually designed post-processing techniques and surpasses state-of-the-art methods, achieving 49.3% AP on MHPv2.0 and 60.4% AP on CIHP. We have released our source code, pretrained models, and demos to facilitate future studies on https://github.com/cjm-sfw/Uniparser.

Authors

Jiaming Chu

3 papers

Lei Jin

1 papers

Yinglei Teng

1 papers

TL;DR

Abstract

Authors

References54 items

Single-Stage Multi-human Parsing via Point Sets and Center-based Offsets

Contextual Instance Decoupling for Instance-Level Human Analysis

Learning Equivariant Segmentation with Instance-Unique Querying

RepParser: End-to-End Multiple Human Parsing with Representative Parts

AIParsing: Anchor-Free Instance-Level Human Parsing

Single-Stage is Enough: Multi-Person Absolute 3D Pose Estimation

Region-Aware Metric Learning for Open World Semantic Segmentation via Meta-Channel Aggregation

Distilling Inter-Class Distance for Semantic Segmentation

Rethinking Semantic Segmentation: A Prototype View

Locality-Aware Inter-and Intra-Video Reconstruction for Self-Supervised Correspondence Learning

Self-supervised Semantic Segmentation Grounded in Visual Concepts

Pose-Guided Hierarchical Semantic Decomposition and Composition for Human Parsing

IMENet: Joint 3D Semantic Scene Completion and 2D Semantic Segmentation through Iterative Mutual Enhancement

Research on Semantic Analysis Strategy in Intelligent Human-Computer Interaction System

Multi-human Parsing with a Graph-based Generative Adversarial Model

Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression

Differentiable Multi-Granularity Human Representation Learning for Instance-Aware Human Semantic Parsing

Exploring Cross-Image Pixel Contrast for Semantic Segmentation

Renovating Parsing R-CNN for Accurate Multiple Human Parsing

Boundary-preserving Mask R-CNN

End-to-End Object Detection with Transformers

SOLOv2: Dynamic and Fast Instance Segmentation

Conditional Convolutions for Instance Segmentation

Learning Semantic Neural Tree for Human Parsing

SOLO: Segmenting Objects by Locations

Self-Correction for Human Parsing

PolarMask: Single Shot Instance Segmentation With Polar Representation

Single-Stage Multi-Person Pose Machines

Zhēng 筝

MMDetection: Open MMLab Detection Toolbox and Benchmark

Fine-Grained Multi-human Parsing

YOLACT: Real-Time Instance Segmentation

Parsing R-CNN for Instance-Level Human Analysis

Devil in the Details: Towards Accurate Single and Multiple Human Parsing

Instance-level Human Parsing via Part Grouping Network

Understanding Humans in Crowded Scenes: Deep Nested Adversarial Learning and A New Benchmark for Multi-Human Parsing

Focal Loss for Dense Object Detection

Body Physical Measurement Using Kinect for Vitual Dressing Room

Joint Multi-person Pose Estimation and Semantic Part Segmentation

Towards Real World Human Parsing: Multiple-Human Parsing in the Wild

Multiple-Human Parsing in the Wild

Semantic Instance Segmentation via Deep Metric Learning

Mask R-CNN

Feature Pyramid Networks for Object Detection

V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation

Deep Residual Learning for Image Recognition

Deep Level Situation Understanding for Casual Communication in Humans-Robots Interaction

Human Computer Interaction for Vision Based Hand Gesture Recognition: A Survey

ReSParser: Fully Convolutional Multiple Human Parsing With Representative Sets

Virtual Glasses Try-on Based on Large Pose Estimation

the B.S. degree from Shandong University, China

Nds Lab -networking and Distributed Systems User Activity Recognition via Kinect in an Ambient Intelligence Scenario User Activity Recognition via Kinect in an Ambient Intelligence Scenario

such as human action recognition, instance segmentation, and human parsing

YunchaoWeiiswithBeijing Jiaotong University, Beijing 100044, China

Field of Study

Journal Information

Name

Page

Volume

Venue Information

Name

Type

URL

Alternate Names