Jiun Tian HOE

West, Singapore

My name is Hoe Jiun Tian (何俊添). I am graduating PhD candidate from Nanyang Technological University, Singapore under supervision of Prof Tan Yap Peng and Prof Jiang Xudong. I earned my Bachelor’s degree in Computer Science (Artificial Intelligence) from the University of Malaya, where I conducted undergraduate research under the supervision of Prof Chan Chee Seng.

My research focuses on computer vision and deep learning, with a specializing in generative AI, particularly for image and video generation.

Image Generation/Editing - enable generative models to move beyond object placement toward realistic, interaction-driven scene composition - InteractDiffusion (CVPR’24), InteractEdit, OneHOI (CVPR’26)
Exocentric to Egocentric Video Generation - transforms exocentric view video to egocentric (first-person) view - TPDF, DRSA, ARPE, PCSE
Deep Hashing-based Image Retrieval - OrthoHash (NeurIPS’21), SDC

Please feel free to reach out directly or view my CV here.

selected publications

CVPR

OneHOI: Unifying Human-Object Interaction Generation and Editing

Jiun Tian Hoe, Weipeng Hu, Xudong Jiang, and 2 more authors

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

Bib HTML Code

@inproceedings{hoe2026onehoi,
  title = {OneHOI: Unifying Human-Object Interaction Generation and Editing},
  author = {Hoe, Jiun Tian and Hu, Weipeng and Jiang, Xudong and Tan, Yap-Peng and Chan, Chee Seng},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2026},
}

Neurocomputing
Reference-free human-object interaction editing

Jiun Tian Hoe, Weipeng Hu, Wei Zhou, and 5 more authors

Neurocomputing, 2026

Abs DOI Bib HTML Code

This paper presents InteractEdit, a novel framework for reference-free Human-Object Interaction (HOI) editing that tackles the challenging task of transforming an existing interaction in an image into a new, desired interaction while preserving the identities of the subject and object. Unlike prior image editing tasks such as attribute manipulation, object replacement or style transfer, HOI editing involves complex spatial, contextual, and relational dependencies inherent in HOI. Existing methods often overfit to the source image structure, limiting adaptability to the substantial structural modifications demanded by the new interactions. To address this, InteractEdit disassembles each scene into subject, object, and background components to disentangle intricate HOI relationships, and introduces a selective inversion strategy combined with Selective-Rank Adaptation (SeRA) to leverage pretrained interaction priors while learning visual identity from the source image. This enables a balanced trade-off between interaction editing and identity preservation. We also introduce IEBench, a new benchmark for HOI editing, and a new metric that jointly evaluates the trade-off between successful interaction editing and identity preservation. Extensive experiments show that InteractEdit outperforms 23 existing methods, providing a strong baseline for future HOI editing research. Code and the dataset: https://jiuntian.github.io/InteractEdit/.
@article{HOE2026134869, title = {Reference-free human-object interaction editing}, journal = {Neurocomputing}, volume = {704}, pages = {134869}, year = {2026}, issn = {0925-2312}, doi = {https://doi.org/10.1016/j.neucom.2026.134869}, url = {https://www.sciencedirect.com/science/article/pii/S0925231226022678}, author = {Hoe, Jiun Tian and Hu, Weipeng and Zhou, Wei and Xie, Chao and Wang, Ziwei and Jiang, Xudong and Tan, Yap-Peng and Chan, Chee Seng}, keywords = {Human-object interaction, Image editing, Diffusion models, Generative models, Deep learning, Computer vision} }

CVPR

InteractDiffusion: Interaction Control in Text-to-Image Diffusion Models

Jiun Tian Hoe, Xudong Jiang, Chee Seng Chan, and 2 more authors

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

Awards Bib HTML PDF Video Code

PREMIA’24: PREMIA Best Student Paper Awards 2024 Honourable Mention Awards

@inproceedings{interactdiffusion2024,
  title = {InteractDiffusion: Interaction Control in Text-to-Image Diffusion Models},
  author = {Hoe, Jiun Tian and Jiang, Xudong and Chan, Chee Seng and Tan, Yap-Peng and Hu, Weipeng},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2024},
  abs = {Large-scale text-to-image (T2I) diffusion models have showcased incredible capabilities in generating coherent images based on textual descriptions, enabling vast applications in content generation. While recent advancements have introduced control over factors such as object localization, posture, and image contours, a crucial gap remains in our ability to control the interactions between objects in the generated content. Well-controlling interactions in generated images could yield meaningful applications, such as creating realistic scenes with interacting characters. In this work, we study the problems of conditioning T2I diffusion models with Human-Object Interaction (HOI) information, consisting of a triplet label (person, action, object) and corresponding bounding boxes. We propose a pluggable interaction control model, called InteractDiffusion that extends existing pre-trained T2I diffusion models to enable them being better conditioned on interactions. Specifically, we tokenize the HOI information and learn their relationships via interaction embeddings. A conditioning self-attention layer is trained to map HOI tokens to visual tokens, thereby conditioning the visual tokens better in existing T2I diffusion models. Our model attains the ability to control the interaction and location on existing T2I diffusion models, which outperforms existing baselines by a large margin in HOI detection score, as well as fidelity in FID and KID. Project page: https://jiuntian.github.io/interactdiffusion.},
  url = {https://arxiv.org/pdf/2312.05849.pdf},
}

NeurIPS

One Loss for All: Deep Hashing with a Single Cosine Similarity based Learning Objective

Jiun Tian Hoe, Kam Woh Ng, Tianyu Zhang, and 3 more authors

In Advances in Neural Information Processing Systems (NeurIPS), 2021

Awards Bib HTML PDF Video Code

IEEE Malaysia FYP Competition 2021 Champion in Signal and Image Processing and Analysis track

@inproceedings{orthohash2021,
  title = {One Loss for All: Deep Hashing with a Single Cosine Similarity based Learning Objective},
  author = {Hoe, Jiun Tian and Ng, Kam Woh and Zhang, Tianyu and Chan, Chee Seng and Song, Yi-Zhe and Xiang, Tao},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2021},
  abs = {A deep hashing model typically has two main learning objectives: to make the learned binary hash codes discriminative and to minimize a quantization error. With further constraints such as bit balance and code orthogonality, it is not uncommon for existing models to employ a large number (>4) of losses. This leads to difficulties in model training and subsequently impedes their effectiveness. In this work, we propose a novel deep hashing model with only . Specifically,  we show that maximizing the cosine similarity between the continuous codes and their corresponding  can ensure both hash code discriminativeness and quantization error minimization. Further, with this learning objective, code balancing can be achieved by simply using a  Batch Normalization (BN) layer and multi-label classification is also straightforward with label smoothing. The result is a one-loss deep hashing model that removes all the hassles of tuning the weights of various losses. Importantly,  extensive experiments show that our model is highly effective, outperforming the state-of-the-art multi-loss hashing models on three large-scale instance retrieval benchmarks, often by significant margins. },
  url = {https://proceedings.neurips.cc/paper/2021/hash/cbcb58ac2e496207586df2854b17995f-Abstract.html},
}