About me

I am a third-year PhD candidate in the Computer Vision Department at Mohamed Bin Zayed University of Artificial Intelligence advised by Prof. Fahad Khan, where I work on designing accurate, fast, and memory-efficient computer vision architectures for edge devices. My PhD CGPA is 3.95 (A).

I have a mixed experience between conducting pure academic research and contributing to international products in the industry. I worked as a Machine Learning Engineer at Valeo Egypt and Teaching/Lecturer Assistant at Faculty of Computer and Information Sciences, Ain Shams University. During my PhD, I strive to build state-of-the-art methods that are efficient, fast, robust, and reliable that can be used for mobile vision applications.

Email  /  Resume  /  Google Scholar  /  Github

Research

Self-attention has become a defacto choice for capturing global context in various vision applications. However, its quadratic computational complexity with respect to image resolution limits its use in real-time applications, especially for deployment on resource-constrained mobile devices. Although hybrid approaches have been proposed to combine the advantages of convolutions and self-attention for a better speed-accuracy trade-off, the expensive matrix multiplication operations in self-attention remain a bottleneck. In this work, we introduce a novel efficient additive attention mechanism that effectively replaces the quadratic matrix multiplication operations with linear element-wise multiplications. Our design shows that the key-value interaction can be replaced with a linear layer without sacrificing any accuracy. Unlike previous state-of-the-art methods, our efficient formulation of self-attention enables its usage at all stages of the network. Using our proposed efficient additive attention, we build a series of models called "SwiftFormer" which achieves state-of-the-art performance in terms of both accuracy and mobile inference speed. Our small variant achieves 78.5% top-1 ImageNet-1K accuracy with only 0.8~ms latency on iPhone 14, which is more accurate and 2x faster compared to MobileViT-v2.

In this paper, we propose a 3D medical image segmentation approach, named UNETR++, that offers both high-quality segmentation masks as well as efficiency in terms of parameters and compute cost. The core of our design is the introduction of a novel efficient paired attention (EPA) block that efficiently learns spatial and channel-wise discriminative features using a pair of inter-dependent branches based on spatial and channel attention. Our spatial attention formulation is efficient having linear complexity with respect to the input sequence length. To enable communication between spatial and channel-focused branches, we share the weights of query and key mapping functions that provide a complimentary benefit (paired attention), while also reducing the overall network parameters. Our extensive evaluations on three benchmarks, Synapse, BTCV and ACDC, reveal the effectiveness of the proposed contributions in terms of both efficiency and accuracy. On Synapse dataset, our UNETR++ sets a new state-of-the-art with a Dice Similarity Score of 87.2%, while being significantly efficient with a reduction of over 71% in terms of both parameters and FLOPs, compared to the best existing method in the literature.

We present EdgeNeXt, a new hybrid architecture that effectively combine the strengths of both CNN and Transformer models. Specifically in EdgeNeXt, we introduce split depth-wise transpose attention (SDTA) encoder that splits input tensors into multiple channel groups and utilizes depth-wise convolution along with self-attention across channel dimensions to implicitly increase the receptive field and encode multi-scale features. Our extensive experiments on classification, detection and segmentation tasks, reveal the merits of the proposed approach, outperforming state-of-the-art methods with comparatively lower compute requirements. Our EdgeNeXt model with 1.3M parameters achieves 71.2% top-1 accuracy on ImageNet-1K, outperforming MobileViT with an absolute gain of 2.2% with 28% reduction in FLOPs. Further, our EdgeNeXt model with 5.6M parameters achieves 81.1% (with knowledge distillation) and 79.4% (without knowledge distillation) top-1 accuracy on ImageNet-1K.

We propose Insta-YOLO, a novel one-stage end-to-end deep learning model for real-time instance segmentation. Instead of pixel-wise prediction, our model predicts instances as object contours represented by 2D points in Cartesian space. We evaluate our model on three datasets, namely, Carvana,Cityscapes and Airbus. We compare our results to the state-of-the-art models for instance segmentation. The results show our model achieves competitive accuracy in terms of mAP at twice the speed on GTX-1080 GPU

We propose a novel data-augmentation technique based on generative adversarial networks (GANs) to restore the balance of the MITBIH dataset. Then, two deep learning approaches—an end-to-end approach and a two-stage hierarchical approach—based on deep convolutional neural networks (CNNs) are used for heartbeat classification. Results show that augmenting the original imbalanced dataset with generated heartbeats by using the proposed techniques more effectively improves the performance of ECG classification than using the same techniques trained only with the original dataset. Furthermore, we demonstrate that augmenting the heartbeats using GANs outperforms other common data augmentation techniques.

News

Publications

Academic Service

Teaching