Mirror-Aware Neural Humans (3DV 2024)
Learning human body pose, shape, and appearance from mirror images and 2D detections.
Biography
I am a Ph.D student in Computer Science at The University of British Columbia (UBC) working with Prof Helge Rhodin and Prof Andrea Tagliasacchi, with interest on neural body models, human motion learning and 3D reconstruction.
I was previously a Google and Facebook scholar at the African Masters in Machine Intelligence (AMMI) program. I am blessed to have worked with Dr. Gaurav Bharaj and Dr. Pablo Garrido at Flawless AI, mentored by Dr. Gamaleldin Elsayed from Google Brain, and taught by Prof. Leonid Sigal at UBC.
News
- Our work on "Mirror-Aware Neural Humans" 🤩 received the best paper award at the weakly supervised computer vision workshop at Deep Learning Indaba 2023 and a poster prize 🏆 at the main event. Also accepted to 3DV 2024!
- I am so excited and grateful to co-organize and lead the first member-led Black in AI social at CVPR 2023.
- I graduated and completed my MSc. program at The University of British Columbia!
Recent Work
My work has largely focused on solving problems in human motion learning, 3D reconstruction, self-supervised learning, and generative body models.
Interaction Design for VR Applications (CHI 2022)
Understanding Needs for University Curricula.
Physical Distancing
Estimating the metric distance between persons from a single video.
3D Human Pose Estimation (NeurIPS DiffCVGP 2020)
An End-End integral pose regression for 2D and 3D Human pose estimation.
Instance Segmentation on a custom dataset with Mask RCNN
Instance segmentation was carried out on a nuts custom dataset that was trained on an object detection library (Detectron2).
Building an Object Tracker
A simple object tracking system that detects objects on all video frames using predictions from a detectron model and links these detections across the temporal domain.
Facial Keypoint Detection
Estimating the location of 68 distinguishing keypoints on human faces.
Automatic Image Captioning
Automatically generating captions that describe the content in images.
Real-Time-Object-detection-API
Implementing a real-time object detection with Tensorflow and OpenCV.
Landmark Detection & Robot Tracking (SLAM)
Implementing SLAM (Simultaneous Localization and Mapping) for a 2 dimensional world!.
Publications
- Ajisafe D.*, Tang J., Su S., Wandt B. and Rhodin H. "Mirror-Aware Neural Humans" | Paper | 3DV 2024
- Ajisafe D.*, Domguia J., Tang J., Ellenberger L., Spörri J. and Rhodin H. "Towards end-to-end training of proposal-based 3D human pose estimation" | Paper | NeurIPS DiffCVGP 2020 | Slideslive
- Biermann O.*, Ajisafe D.*, Yoon D*. "Interaction for VR Applications: Understanding Needs for University Curricula" | Paper | CHI 2022 | DOI
Community-based Publications
- Ogundepo O.*, et., al. AfriQA: Cross-lingual Open-Retrieval Question Answering for African Languages | Paper | EMNLP Findings 2023
Invited talks
- I am delighted to give talk on "Masterclass to Fundamentals of Computer Vision and Image Processing"
Host: DSN - Data Scientists Network (Data Science Nigeria) | Youtube - It was such an honor to speak on the “The Future of Artificial Intelligence in Africa and in the Diaspora” at Kampala International University.