Muhammad Kamran Janjua

Art is Long; Time is Fleeting!

I am a Machine Learning Researcher @ Huawei, Canada. I work on video understanding and online learning problems. Previously, I was a graduate student with Dr. Martha White at University of Alberta, where I worked on continual learning.

Email  /  CV  /  Google Scholar  /  Github

I am interested in problems in perception and learning, particularly in learning from videos. Some (representative) papers are highlighted.

We propose an all-in-one video restoration framework that grounds degradation-aware semantic context of video frames in natural language via foundation models, offering interpretable and flexible guidance.

We propose ConicGrad, a principled, scalable, and robust MTL approach formulated as a constrained optimization problem. Our method introduces an angular constraint to dynamically regulate gradient update directions, confining them within a cone centered on the reference gradient of the overall objective.

We propose Turtle to learn the truncated causal history model for online video processing. The causal design in Turtle enables recurrence in inference through state-memorized historical features while allowing parallel training by sampling truncated video clips. We report new state-of-the-art results on a multitude of video restoration benchmark tasks.

We present CascadedGaze Network (CGNet), an encoder-decoder architecture that employs Global Context Extractor (GCE), a novel and efficient way to learn global information for image restoration.

We show the importance of learning in deployment, by comparing a TD agent trained purely offline with no online updating to a TD agent that learns online. This final result is one of the first to motivate the importance of adapting predictions in real-time, for non-stationary high-volume systems in the real world.

We propose a method for fusing stereo disparity estimation with movement-induced prior information.

We propose to learn a joint representation of audio and visual information without relying on multimodal pairs or triplets for matching, verficiation, and retrieval tasks.

We propose a new measure SemanticMap to evaluate the performance of cross modal systems. Our proposed measure evaluates the semantic similarity between the image and text representations in the latent embedding space.

In order to further enhance the discriminative capability of deep features, we introduce a joint supervision signal, Git loss, which leverages on softmax and center loss functions. The aim of our loss function is to minimize the intra-class variations as well as maximize the inter-class distances.