mlsanityData to 7 Oct 2026

Parallel Multi Channel Convolution using General Matrix Multiplication

arXiv 1704.04428 ↗First posted 6 Apr 2017cs.CV cs.PFcited by 30

Not on the map. Why?

Abstract (arXiv)

Convolutional neural networks (CNNs) have emerged as one of the most successful machine learning technologies for image and video processing. The most computationally intensive parts of CNNs are the convolutional layers, which convolve multi-channel images with multiple kernels. A common approach to implementing convolutional layers is to expand the image into a column matrix (im2col) and perform Multiple Channel Multiple Kernel (MCMK) convolution using an existing parallel General Matrix Multiplication (GEMM) library. This im2col conversion greatly increases the memory footprint of the input matrix and reduces data locality. In this paper we propose a new approach to MCMK convolution that is based on General Matrix Multiplication (GEMM), but not on im2col. Our algorithm eliminates the need for data replication on the input thereby enabling us to apply the convolution kernels on the input images directly. We have implemented several variants of our algorithm on a CPU processor and an embedded ARM processor. On the CPU, our algorithm is faster than im2col in most cases.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Apr 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 6; 2022: 6; 2023: 5; 2024: 7; 2025: 3; 2026: 3.

30 papers.

Sort:

By area:

  1. 2022GradMax: Growing Neural Networks using Gradient Informationcited by 64
  2. 2021Fast convolutional neural networks on FPGAs with hls4mlcited by 42
  3. 2021Neuro-Symbolic AI: An Emerging Class of AI Workloads and their Characterizationcited by 22
  4. 2021Post-Training Sparsity-Aware Quantizationcited by 16
  5. 2025Systems and Algorithms for Convolutional Multi-Hybrid Language Models at Scalecited by 12
  6. 2024When Linear Attention Meets Autoregressive Decoding: Towards More Effective and Efficient Linearized Large Language Modelscited by 9
  7. 2023All Rivers Run to the Sea: Private Learning with Asymmetric Flowscited by 4
  8. 2021SMASH: Sparse Matrix Atomic Scratchpad Hashingcited by 3
  9. 2021Content-Aware Convolutional Neural Networkscited by 3
  10. 2023Advancing Direct Convolution using Convolution Slicing Optimization and ISA Extensionscited by 3
  11. 2024FairViT: Fair Vision Transformer via Adaptive Maskingcited by 3
  12. 2024BF-IMNA: A Bit Fluid In-Memory Neural Architecture for Neural Network Accelerationcited by 3
18 more
  1. 2023MEMA Runtime Framework: Minimizing External Memory Accesses for TinyML on Microcontrollerscited by 2
  2. 2024Tazza: Shuffling Neural Network Parameters for Secure and Private Federated Learningcited by 2
  3. 2021Efficient and Generic 1D Dilated Convolution Layer for Deep Learningcited by 1
  4. 2022Analytic Learning of Convolutional Neural Network For Pattern Recognitioncited by 1
  5. 2022OLLIE: Derivation-based Tensor Program Optimizercited by 1
  6. 2023Sliding Window Sum Algorithms for Deep Neural Networkscited by 1
  7. 2024TabConv: Low-Computation CNN Inference via Table Lookupscited by 1
  8. 2025FairLRF: Achieving Fairness through Sparse Low Rank Factorizationcited by 1
  9. 2026Theory of Minimal Weight Perturbations in Deep Networks and its Applications for Low-Rank Activated Backdoor Attackscited by 1
  10. 2022Lipschitz Bound Analysis of Neural Networkscited by 0
  11. 2022Fast-ParC: Capturing Position Aware Global Feature for ConvNets and ViTscited by 0
  12. 2022Towards a learning-based performance modeling for accelerating Deep Neural Networkscited by 0
  13. 2023Accelerating Machine Learning Primitives on Commodity Hardwarecited by 0
  14. 2024Optimizing Sparse Convolution on GPUs with CUDA for 3D Point Cloud Processing in Embedded Systemscited by 0
  15. 2024A Framework to Enable Algorithmic Design Choice Exploration in DNNscited by 0
  16. 2025Unified Kernel-Segregated Transpose Convolution Operationcited by 0
  17. 2026Reducing the Complexity of Matrix Multiplication by Quantum Computingcited by 0
  18. 2026ALADIN: Accuracy-Latency-Aware Design-space Inference Analysis for Embedded AI Acceleratorscited by 0