Search Results - "Crowley, Elliot J."

Substituting Convolutions for Neural Network Compression by Crowley, Elliot J., Gray, Gavin, Turner, Jack, Storkey, Amos

“…Many practitioners would like to deploy deep, convolutional neural networks in memory-limited scenarios, e.g., on an embedded device. However, with an…”
Get full text

Optimizing Grouped Convolutions on Edge Devices by Gibson, Perry, Cano, Jose, Turner, Jack, Crowley, Elliot J., O'Boyle, Michael, Storkey, Amos

“…When deploying a deep neural network on con-strained hardware, it is possible to replace the network's standard convolutions with grouped convolutions. This…”
Get full text

Performance Aware Convolutional Neural Network Channel Pruning for Embedded GPUs by Radu, Valentin, Kaszyk, Kuba, Wen, Yuan, Turner, Jack, Cano, Jose, Crowley, Elliot J., Franke, Bjorn, Storkey, Amos, O'Boyle, Michael

“…Convolutional Neural Networks (CNN) are becoming a common presence in many applications and services, due to their superior recognition accuracy. They are…”
Get full text

Generate Your Own Scotland: Satellite Image Generation Conditioned on Maps by Espinosa, Miguel, Crowley, Elliot J

“…Despite recent advancements in image generation, diffusion models still remain largely underexplored in Earth Observation. In this paper we show that…”
Get full text

Plug and Play Active Learning for Object Detection by Yang, Chenhongyi, Huang, Lichao, Crowley, Elliot J.

“…Annotating datasets for object detection is an expensive and time-consuming endeavor. To minimize this burden, active learning (AL) techniques are employed to…”
Get full text

Plug and Play Active Learning for Object Detection by Yang, Chenhongyi, Huang, Lichao, Crowley, Elliot J

“…Annotating datasets for object detection is an expensive and time-consuming endeavor. To minimize this burden, active learning (AL) techniques are employed to…”
Get full text

WidthFormer: Toward Efficient Transformer-based BEV View Transformation by Yang, Chenhongyi, Lin, Tianwei, Huang, Lichao, Crowley, Elliot J

“…We present WidthFormer, a novel transformer-based module to compute Bird's-Eye-View (BEV) representations from multi-view cameras for real-time…”
Get full text

Contrastive Object-level Pre-training with Spatial Noise Curriculum Learning by Yang, Chenhongyi, Huang, Lichao, Crowley, Elliot J

“…The goal of contrastive learning based pre-training is to leverage large quantities of unlabeled data to produce a model that can be readily adapted…”
Get full text

Neural Architecture Search as Program Transformation Exploration by Turner, Jack, Crowley, Elliot J, O'Boyle, Michael

“…Improving the performance of deep neural networks (DNNs) is important to both the compiler and neural architecture search (NAS) communities. Compilers apply…”
Get full text

Hyperparameter Selection in Continual Learning by Lee, Thomas L, Hellan, Sigrid Passano, Ericsson, Linus, Crowley, Elliot J, Storkey, Amos

“…In continual learning (CL) -- where a learner trains on a stream of data -- standard hyperparameter optimisation (HPO) cannot be applied, as a learner does not…”
Get full text

EgoPoseFormer: A Simple Baseline for Stereo Egocentric 3D Human Pose Estimation by Yang, Chenhongyi, Tkach, Anastasia, Hampali, Shreyas, Zhang, Linguang, Crowley, Elliot J, Keskin, Cem

“…We present EgoPoseFormer, a simple yet effective transformer-based model for stereo egocentric human pose estimation. The main challenge in egocentric pose…”
Get full text

PlainMamba: Improving Non-Hierarchical Mamba in Visual Recognition by Yang, Chenhongyi, Chen, Zehui, Espinosa, Miguel, Ericsson, Linus, Wang, Zhenyu, Liu, Jiaming, Crowley, Elliot J

“…We present PlainMamba: a simple non-hierarchical state space model (SSM) designed for general visual recognition. The recent Mamba model has shown how SSMs can…”
Get full text

Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks by Turner, Jack, Cano, Jose, Radu, Valentin, Crowley, Elliot J., O'Boyle, Michael, Storkey, Amos

“…Convolutional Neural Networks (CNNs) are extremely computationally demanding, presenting a large barrier to their deployment on resource-constrained devices…”
Get full text

Prediction-Guided Distillation for Dense Object Detection by Yang, Chenhongyi, Ochal, Mateusz, Storkey, Amos, Crowley, Elliot J

“…Real-world object detection models should be cheap and accurate. Knowledge distillation (KD) can boost the accuracy of a small, cheap detection model by…”
Get full text

DLAS: An Exploration and Assessment of the Deep Learning Acceleration Stack by Gibson, Perry, Cano, José, Crowley, Elliot J, Storkey, Amos, O'Boyle, Michael

“…Deep Neural Networks (DNNs) are extremely computationally demanding, which presents a large barrier to their deployment on resource-constrained devices. Since…”
Get full text

einspace: Searching for Neural Architectures from Fundamental Operations by Ericsson, Linus, Espinosa, Miguel, Yang, Chenhongyi, Antoniou, Antreas, Storkey, Amos, Cohen, Shay B, McDonagh, Steven, Crowley, Elliot J

“…Neural architecture search (NAS) finds high performing networks for a given task. Yet the results of NAS are fairly prosaic; they did not e.g. create a shift…”
Get full text

GPViT: A High Resolution Non-Hierarchical Vision Transformer with Group Propagation by Yang, Chenhongyi, Xu, Jiarui, De Mello, Shalini, Crowley, Elliot J, Wang, Xiaolong

“…We present the Group Propagation Vision Transformer (GPViT): a novel nonhierarchical (i.e. non-pyramidal) transformer model designed for general visual…”
Get full text

Separable Layers Enable Structured Efficient Linear Substitutions by Gray, Gavin, Crowley, Elliot J, Storkey, Amos

“…In response to the development of recent efficient dense layers, this paper shows that something as simple as replacing linear components in pointwise…”
Get full text

Neural Architecture Search without Training by Mellor, Joseph, Turner, Jack, Storkey, Amos, Crowley, Elliot J

“…The time and effort involved in hand-designing deep neural networks is immense. This has prompted the development of Neural Architecture Search (NAS)…”
Get full text

Moonshine: Distilling with Cheap Convolutions by Crowley, Elliot J, Gray, Gavin, Storkey, Amos

“…Many engineers wish to deploy modern neural networks in memory-limited settings; but the development of flexible methods for reducing memory use is in its…”
Get full text