Amirul Islam
Toronto, May 2020 |
I am a machine learning researcher with primary interests in generative AI, multi-modal learning, and human-centric AI (e.g., explainability, interpretability, generalization, fairness, and bias). Previously, I was a senior researcher at Huawei Noah's Ark Lab in Toronto. I received my Ph.D. in Computer Science from Ryerson University in 2022, advised by Neil Bruce and Kosta Derpanis. Prior to that, I completed my M.Sc. in Computer Science at the University of Manitoba, advised by Yang Wang and Neil Bruce. I am also a recipient of the Vector Institute Postgraduate Affiliate award. Email: amirul507 [at] gmail.com |
News
- Sep 2024 » One paper accepted to IEEE TPAMI!
- Mar 2024 » One paper accepted to IJCV.
- Oct 2023 » One paper accepted to WACV 2024 as Oral!
- Feb 2023 » One paper accepted to IJCV.
- Feb 2022 » One paper accepted to CVPR 2022
- Jan 2022 » I graduated with a PhD in Computer Science from Ryerson University.
Publications & Preprints
Visually Guided Audio Source Separation with Meta Consistency Learning
Quantifying and Learning Static vs. Dynamic Information in Deep Spatiotemporal Networks
Position, Padding and Predictions: A Deeper Look at Position Information in CNNs
SegMix: Co-occurrence Driven Mixup for Semantic Segmentation and Adversarial Robustness
Maximizing Mutual Shape Information
A Deeper Dive Into What Deep Spatiotemporal Networks Encode: Quantifying Static vs. Dynamic Information
Global Pooling, More than Meets the Eye: Position Information is Encoded Channel-Wise in CNNs
Simpler Does It: Generating Semantic Labels with Objectness Guidance
Shape or Texture: Understanding Discriminative Features in CNNs
Bidirectional Attention Network for Monocular Depth Estimation
Fibro-CoSANet: Pulmonary Fibrosis Prognosis Prediction using a
Convolutional Self Attention Network
Feature Binding with Category-Dependant MixUp for Semantic Segmentation and Adversarial Robustness
How much Position Information Do Convolutional Neural Networks Encode?
Distributed Iterative Gating Networks for Semantic Segmentation
Relative Saliency and Ranking: Models, Metrics, Data and Benchmarks
Recurrent Iterative Gating Networks for Semantic Segmentation
Revisiting Salient Object Detection: Simultaneous Detection, Ranking, and Subitizing of Multiple Salient Objects
Semantics Meet Saliency: Exploring Domain Affinity and Models for Dual-Task Prediction
Gated Feedback Refinement Network for Coarse-to-Fine Dense Semantic Image Labeling
Gated Feedback Refinement Network for Dense Image Labeling
On the Robustness of Deep Learning Models to Universal Adversarial Attack
Salient Object Detection using a Context-Aware Refinement Network
Label Refinement Network for Coarse-to-Fine Semantic Segmentation
Dense Image Labeling Using Deep Convolutional Neural Networks