SOTAVerified

Human Parsing

Human parsing is the task of segmenting a human image into different fine-grained semantic parts such as head, torso, arms and legs.

( Image credit: Multi-Human-Parsing (MHP) )

Papers

Showing 110 of 125 papers

TitleStatusHype
Beyond Appearance: a Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual TasksCode3
4D-DRESS: A 4D Dataset of Real-world Human Clothing with Semantic AnnotationsCode2
HumanBench: Towards General Human-centric Perception with Projector Assisted PretrainingCode2
Text2Human: Text-Driven Controllable Human Image GenerationCode2
InvPT: Inverted Pyramid Multi-task Transformer for Dense Scene UnderstandingCode2
Body Part-Based Representation Learning for Occluded Person Re-IdentificationCode2
Correlating Edge, Pose with ParsingCode1
End-to-end One-shot Human ParsingCode1
CDGNet: Class Distribution Guided Network for Human ParsingCode1
Differentiable Multi-Granularity Human Representation Learning for Instance-Aware Human Semantic ParsingCode1
Show:102550
← PrevPage 1 of 13Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Graphonomy_InnermAcc0.97Unverified
2CDGNet_InnermAcc0.92Unverified
3Graphonomy_OutermAcc0.92Unverified
4SCHP_InnermAcc0.91Unverified
5CDGNet_OutermAcc0.89Unverified
6SCHP_OutermAcc0.86Unverified
#ModelMetricClaimedVerifiedStatus
1InvPTmIoU67.61Unverified