SOTAVerified

Pedestrian Attribute Recognition

Pedestrian attribution recognition is the task of recognizing pedestrian features - such as whether they are talking on a phone, whether they have a backpack, and so on.

( Image credit: HydraPlus-Net: Attentive Deep Features for Pedestrian Analysis )

Papers

Showing 125 of 56 papers

TitleStatusHype
FOCUS: Fine-grained Optimization with Semantic Guided Understanding for Pedestrian Attributes Recognition0
ViTA-PAR: Visual and Textual Attribute Alignment with Attribute Prompting for Pedestrian Attribute RecognitionCode0
Adversarial Semantic and Label Perturbation Attack for Pedestrian Attribute RecognitionCode0
RGB-Event based Pedestrian Attribute Recognition: A Benchmark Dataset and An Asymmetric RWKV Fusion FrameworkCode0
AttriVision: Advancing Generalization in Pedestrian Attribute Recognition using CLIP0
Enhanced Visual-Semantic Interaction with Tailored Prompts for Pedestrian Attribute Recognition0
SNN-PAR: Energy Efficient Pedestrian Attribute Recognition via Spiking Neural NetworksCode0
Pedestrian Attribute Recognition: A New Benchmark Dataset and A Large Language Model Augmented FrameworkCode0
An Empirical Study of Mamba-based Pedestrian Attribute RecognitionCode0
Pedestrian Attribute Recognition as Label-balanced Multi-label LearningCode1
Spatio-Temporal Side Tuning Pre-trained Foundation Models for Video-based Pedestrian Attribute RecognitionCode1
C2T-Net: Channel-Aware Cross-Fused Transformer-Style Networks for Pedestrian Attribute RecognitionCode1
Pedestrian Attribute Recognition via CLIP based Prompt Vision-Language FusionCode0
SSPNet: Scale and Spatial Priors Guided Generalizable and Interpretable Pedestrian Attribute RecognitionCode0
SequencePAR: Understanding Pedestrian Attributes via A Sequence Generation ParadigmCode0
Hulk: A Universal Knowledge Translator for Human-Centric TasksCode2
HAP: Structure-Aware Masked Image Modeling for Human-Centric PerceptionCode1
A Solution to Co-occurrence Bias: Attributes Disentanglement via Mutual Information Minimization for Pedestrian Attribute RecognitionCode0
Lightweight Attribute Localizing Models for Pedestrian Attribute Recognition0
Knowledge Assembly: Semi-Supervised Multi-Task Learning from Multiple Datasets with Disjoint Labels0
Towards Unified Text-based Person Retrieval: A Large-scale Multi-Attribute and Language Search BenchmarkCode1
PLIP: Language-Image Pre-training for Person Representation LearningCode1
Learning CLIP Guided Visual-Text Fusion Transformer for Video-based Pedestrian Attribute RecognitionCode1
PARFormer: Transformer-based Multi-Task Network for Pedestrian Attribute RecognitionCode1
Beyond Appearance: a Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual TasksCode3
Show:102550
← PrevPage 1 of 3Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AttivisionF1 score93.1Unverified
2PATH (Partial FT)Accuracy90.8Unverified
3AttrivisionAccuracy 89.8Unverified
4Hulk(Finetune, ViT-L)Accuracy88.97Unverified
5Hulk(Finetune, ViT-B)Accuracy87.85Unverified
6C2T-NetAccuracy87.2Unverified
7SOLIDERAccuracy86.38Unverified
8UniHCP (finetune)Accuracy86.18Unverified
9APTMAccuracy80.17Unverified
10Label2LabelAccuracy79.23Unverified
#ModelMetricClaimedVerifiedStatus
1UniHCP (FT)Accuracy88.78Unverified
2C2T-NetAccuracy88.2Unverified
3ALM[tang2019Improving] (ICCV19)Accuracy79.52Unverified
4Attribute-Specific LocalizationAccuracy79.52Unverified
5strongbaselineAccuracy79.14Unverified
6HP-netAccuracy76.13Unverified
#ModelMetricClaimedVerifiedStatus
1Hulk(Finetune, ViT-L)Accuracy85.86Unverified
2Hulk(Finetune, ViT-B)Accuracy85.26Unverified
3AttrivisionAccuracy 84.2Unverified
4UniHCP (finetune)Accuracy82.34Unverified
#ModelMetricClaimedVerifiedStatus
1Attribute-Specific LocalizationAccuracy68.17Unverified
2HP-netAccuracy65.39Unverified
#ModelMetricClaimedVerifiedStatus
1DenseNetBackpack63.9Unverified
2ResNetBackpack63.5Unverified
#ModelMetricClaimedVerifiedStatus
1MBNETAccuracy91.13Unverified
#ModelMetricClaimedVerifiedStatus
1AttrivisionAccuracy 83.8Unverified
#ModelMetricClaimedVerifiedStatus
1AttrivisionAccuracy 89.4Unverified