Ashish Sinha

CV/ML Researcher

Toronto, Canada

I am currently an ML Resident at AMII in the Advanced Technology team.

My research interests lie at the intersection of computer vision and machine learning, with applications in healthcare and scientific discovery. I am particularly interested in developing novel algorithms and leveraging recent advances in representation learning, generative modeling, and geometric deep learning to develop robust and interpretable AI systems dedicated to transforming healthcare, particularly in precision medicine and drug discovery.

I hold an MSc (thesis) in Computer Science from Simon Fraser University, where I worked with Prof. Ghassan Hamarneh, and a Bachelor’s in Materials Science from IIT Roorkee.

Previously, I was an ML Researcher at Huawei’s Noah’s Ark Lab, a Risk Analyst at Wells Fargo, a Research Engineer at Preferred Networks, and an intern with Prof. Jonghyun Choi and Prof. Jose Dolz.

If you are interested in my work, have opportunities, or would like to collaborate, please feel free to reach out.

Offline: 📚 Reading · 🎬 Movies · 🏃 Running · 📷 Photography · 🥏 Ultimate

news

Sep 2, 2025	Starting as Machine Learning Resident at Amii
Aug 1, 2025	UnPose is accepted to CoRL ‘25
Jul 3, 2024	Started as ML Researcher at Huawei Noah’s Ark Lab, Toronto

selected publications

2025

CoRL

UnPose: Uncertainty-Guided Diffusion Priors for Zero-Shot Pose Estimation

Zhaodong Jiang, Ashish Sinha, Tongtong Cao, Yuan Ren, Bingbing Liu, and Binbin Xu

In Conference on Robot Learning (CoRL), 2025

Bib PDF Website

@inproceedings{jiang2025unpose,
  title = {UnPose: Uncertainty-Guided Diffusion Priors for Zero-Shot Pose Estimation},
  author = {Jiang, Zhaodong and Sinha, Ashish and Cao, Tongtong and Ren, Yuan and Liu, Bingbing and Xu, Binbin},
  booktitle = {Conference on Robot Learning (CoRL)},
  year = {2025},
}

2024

MedIA

DermSynth3D: Synthesis of in-the-wild annotated dermatology images

Ashish Sinha, Jeremy Kawahara, Arezou Pakzad, Kumar Abhishek, Matthieu Ruthven, Enjie Ghorbel, Anis Kacem, Djamila Aouada, and Ghassan Hamarneh

Medical Image Analysis, 2024

Bib PDF Code Website

@article{sinha2024dermsynth3d,
  title = {DermSynth3D: Synthesis of in-the-wild annotated dermatology images},
  author = {Sinha, Ashish and Kawahara, Jeremy and Pakzad, Arezou and Abhishek, Kumar and Ruthven, Matthieu and Ghorbel, Enjie and Kacem, Anis and Aouada, Djamila and Hamarneh, Ghassan},
  journal = {Medical Image Analysis},
  volume = {95},
  pages = {103145},
  year = {2024},
  publisher = {Elsevier},
}

MICCAI

TrIND: Representing Anatomical Trees by Denoising Diffusion of Implicit Neural Fields

Ashish Sinha and Ghassan Hamarneh

In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2024

Awarded Bib PDF Code

Best Paper runner-up @ SFU CS Industry Day

@inproceedings{sinha2024trind,
  title = {TrIND: Representing Anatomical Trees by Denoising Diffusion of Implicit Neural Fields},
  author = {Sinha, Ashish and Hamarneh, Ghassan},
  booktitle = {International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)},
  volume = {15012},
  year = {2024},
}

2020

JBHI

Multi-scale self-guided attention for medical image segmentation

Ashish Sinha and Jose Dolz

IEEE journal of biomedical and health informatics, 2020

Bib PDF Code

@article{sinha2020multi,
  title = {Multi-scale self-guided attention for medical image segmentation},
  author = {Sinha, Ashish and Dolz, Jose},
  journal = {IEEE journal of biomedical and health informatics},
  year = {2020},
  publisher = {IEEE},
}