# Unseen No More: Unlocking the Potential of CLIP for Generative Zero-shot HOI Detection

arXiv: https://arxiv.org/abs/2408.05974 · HTML: https://mlsanity.com/paper/2408.05974/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 36 of its 53 references that are in the map.
- First posted on arXiv 12 Aug 2024.

## Builds on (most cited first, 30 of 36)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Learning to Prompt for Vision-Language Models](https://mlsanity.com/paper/2109.01134/) (2021)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [Autoencoding beyond pixels using a learned similarity metric](https://mlsanity.com/paper/1512.09300/) (2015)
- [Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification](https://mlsanity.com/paper/2207.09519/) (2022)
- [Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders](https://mlsanity.com/paper/1812.01784/) (2018)
- [Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learners](https://mlsanity.com/paper/2303.02151/) (2023)
- [QPIC: Query-Based Pairwise Human-Object Interaction Detection with Image-Wide Contextual Information](https://mlsanity.com/paper/2103.05399/) (2021)
- [iCAN: Instance-Centric Attention Network for Human-Object Interaction Detection](https://mlsanity.com/paper/1808.10437/) (2018)
- [GEN-VLKT: Simplify Association and Enhance Interaction Understanding for HOI Detection](https://mlsanity.com/paper/2203.13954/) (2022)
- [DRG: Dual Relation Graph for Human-Object Interaction Detection](https://mlsanity.com/paper/2008.11714/) (2020)
- [Reformulating HOI Detection as Adaptive Set Prediction](https://mlsanity.com/paper/2103.05983/) (2021)
- [Efficient Two-Stage Detection of Human-Object Interactions with a Novel Unary-Pairwise Transformer](https://mlsanity.com/paper/2112.01838/) (2021)
- [HOI Analysis: Integrating and Decomposing Human-Object Interaction](https://mlsanity.com/paper/2010.16219/) (2020)
- [Detecting Human-Object Interaction via Fabricated Compositional Learning](https://mlsanity.com/paper/2103.08214/) (2021)
- [CLIP-GEN: Language-Free Training of a Text-to-Image Generator with CLIP](https://mlsanity.com/paper/2203.00386/) (2022)
- [Dont Even Look Once: Synthesizing Features for Zero-Shot Detection](https://mlsanity.com/paper/1911.07933/) (2019)
- [Exploring Predicate Visual Context in Detecting Human-Object Interactions](https://mlsanity.com/paper/2308.06202/) (2023)
- [AvatarCLIP: Zero-Shot Text-Driven Generation and Animation of 3D Avatars](https://mlsanity.com/paper/2205.08535/) (2022)
- [Detecting Human-Object Interactions via Functional Generalization](https://mlsanity.com/paper/1904.03181/) (2019)
- [Improving Zero-Shot Generalization for CLIP with Synthesized Prompts](https://mlsanity.com/paper/2307.07397/) (2023)
- [ConsNet: Learning Consistency Graph for Zero-Shot Human-Object Interaction Detection](https://mlsanity.com/paper/2008.06254/) (2020)
- [Being Comes from Not-being: Open-vocabulary Text-to-Motion Generation with Wordless Training](https://mlsanity.com/paper/2210.15929/) (2022)
- [Human-Object Interaction Detection via Disentangled Transformer](https://mlsanity.com/paper/2204.09290/) (2022)
- [What to look at and where: Semantic and Spatial Refined Transformer for detecting human-object interactions](https://mlsanity.com/paper/2204.00746/) (2022)
- [Detecting Any Human-Object Interaction Relationship: Universal HOI Detector with Spatial Prompt Learning on Foundation Models](https://mlsanity.com/paper/2311.03799/) (2023)

## Built on by (most cited first, 11 of 11)

- [Open-Vocabulary HOI Detection with Interaction-aware Prompt and Concept Calibration](https://mlsanity.com/paper/2508.03207/) (2025)
- [What if Agents Could Imagine? Reinforcing Open-Vocabulary HOI Comprehension through Generation](https://mlsanity.com/paper/2602.11499/) (2026)
- [Differential Contrastive Training for Gaze Estimation](https://mlsanity.com/paper/2502.20128/) (2025)
- [DQEN: Dual Query Enhancement Network for DETR-based HOI Detection](https://mlsanity.com/paper/2508.18896/) (2025)
- [Mitigating Long-Tail Bias in HOI Detection via Adaptive Diversity Cache](https://mlsanity.com/paper/2511.18811/) (2025)
- [Unleashing Multimodal Large Language Models for Training-free HOI Detection in the Wild](https://mlsanity.com/paper/2607.13881/) (2026)
- [Multimodal Graph Network Modeling for Human-Object Interaction Detection with PDE Graph Diffusion](https://mlsanity.com/paper/2509.12554/) (2025)
- [Contextualized Representation Learning for Effective Human-Object Interaction Detection](https://mlsanity.com/paper/2509.12784/) (2025)
- [Generative Human-Object Interaction Detection via Differentiable Cognitive Steering of Multi-modal LLMs](https://mlsanity.com/paper/2512.17640/) (2025)
- [Towards Unconstrained Human-Object Interaction](https://mlsanity.com/paper/2604.14069/) (2026)
- [FineHOI: Part-Aware Dense Representations for Zero-Shot Human-Object Interaction Detection](https://mlsanity.com/paper/2609.05959/) (2026)

## Abstract (arXiv)

Zero-shot human-object interaction (HOI) detector is capable of generalizing to HOI categories even not encountered during training. Inspired by the impressive zero-shot capabilities offered by CLIP, latest methods strive to leverage CLIP embeddings for improving zero-shot HOI detection. However, these embedding-based methods train the classifier on seen classes only, inevitably resulting in seen-unseen confusion for the model during inference. Besides, we find that using prompt-tuning and adapters further increases the gap between seen and unseen accuracy. To tackle this challenge, we present the first generation-based model using CLIP for zero-shot HOI detection, coined HOIGen. It allows to unlock the potential of CLIP for feature generation instead of feature extraction only. To achieve it, we develop a CLIP-injected feature generator in accordance with the generation of human, object and union features. Then, we extract realistic features of seen samples and mix them with synthetic features together, allowing the model to train seen and unseen classes jointly. To enrich the HOI scores, we construct a generative prototype bank in a pairwise HOI recognition branch, and a multi-knowledge prototype bank in an image-wise HOI recognition branch, respectively. Extensive experiments on HICO-DET benchmark demonstrate our HOIGen achieves superior performance for both seen and unseen classes under various zero-shot settings, compared with other top-performing methods. Code is available at: https://github.com/soberguo/HOIGen