AI Paper Atlas

Research map / Benchmarks / HICO-DET Default Full

HICO-DET Default Full leaderboard

32 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on HICO-DET Default Full. Best mAP: 48.16 (UniHOI, AAAI 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

HICO-DET Default Full · mAP

32 papers, each with its best reported number for its own method; higher is better.

#mAP ↑MethodPaperTopic
148.16UniHOIUniHOI: Unified Human-Object Interaction Understanding via Unified Token Space
AAAI 2026 · Table 1
Open-set and few-shot detection
247.71HOI-IDiffAn Image-like Diffusion Method for Human-Object Interaction Detection
CVPR 2025 · Table 1
Autoregressive and latent generation
347.53HORP
Swin-L+CLIP
HORP: Human-Object Relation Priors Guided HOI Detection
CVPR 2025 · Table 4
Open-set and few-shot detection
446.01Ours
Swin-Large
Exploring Pose-Aware Human-Object Interaction via Hybrid Learning
CVPR 2024 · Table 1
Open-set and few-shot detection
544.32PViC w/ H-DETR (Swin-L)
H-DETR Swin-L
Exploring Predicate Visual Context in Detecting of Human-Object Interactions
ICCV 2023 · Table 6
Open-set and few-shot detection
642.54DiffusionHOI
Stable unCLIP
Human-Object Interaction Detection Collaborated with Large Relation-driven Diffusion Models
NeurIPS 2024 · Table 1
Text-to-image diffusion
742.09ContextHOI
ResNet101
ContextHOI: Spatial Context Learning for Human-Object Interaction Detection
AAAI 2025 · Table 1
Open-set and few-shot detection
840.95UniHOI-l
ResNet-101 + BLIP2
Detecting Any Human-Object Interaction Relationship: Universal HOI Detector with Spatial Prompt Learning on Foundation Models
NeurIPS 2023 · Table 1
Open-set and few-shot detection
938.07UniVRD (LiT: ViT-H/14)
LiT ViT-H/14
Unified Visual Relationship Detection with Vision and Language Models
ICCV 2023 · Table 1
Open-set and few-shot detection
1037.41RmLR
ResNet-101
Re-mine, Learn and Reason: Exploring the Cross-modal Semantic Correlations for Language-guided HOI detection
ICCV 2023 · Table 1
Open-set and few-shot detection
22 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.