Compressing CNN Kernels for Videos Using Tucker Decompositions: Towards Lightweight CNN Applications

• URL: http://arxiv.org/abs/2203.07033v1
• Date: Thu, 10 Mar 2022 11:53:53 GMT
• Title: Compressing CNN Kernels for Videos Using Tucker Decompositions: Towards Lightweight CNN Applications
• Authors: Tobias Engelhardt Rasmussen, Line H Clemmensen and Andreas Baum
• Abstract summary: Convolutional Neural Networks (CNN) are the state-of-theart in the field of visual computing. A major problem with CNNs is the large number of floating point operations (FLOPs) required to perform convolutions for large inputs. We propose a Tuckerdecomposition to compress the convolutional kernel of a pre-trained network for images.
• Score: 2.191505742658975
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Convolutional Neural Networks (CNN) are the state-of-the-art in the field of visual computing. However, a major problem with CNNs is the large number of floating point operations (FLOPs) required to perform convolutions for large inputs. When considering the application of CNNs to video data, convolutional filters become even more complex due to the extra temporal dimension. This leads to problems when respective applications are to be deployed on mobile devices, such as smart phones, tablets, micro-controllers or similar, indicating less computational power. Kim et al. (2016) proposed using a Tucker-decomposition to compress the convolutional kernel of a pre-trained network for images in order to reduce the complexity of the network, i.e. the number of FLOPs. In this paper, we generalize the aforementioned method for application to videos (and other 3D signals) and evaluate the proposed method on a modified version of the THETIS data set, which contains videos of individuals performing tennis shots. We show that the compressed network reaches comparable accuracy, while indicating a memory compression by a factor of 51. However, the actual computational speed-up (factor 1.4) does not meet our theoretically derived expectation (factor 6).

Related papers

• Tensor network compressibility of convolutional models [0.0]
  We assess how truncating the convolution kernels of dense (untensorized) CNNs impact their accuracy. We found that kernels could often be truncated along several cuts resulting in significant loss in kernel norm but not in classification accuracy.
  arXiv  Detail & Related papers  (2024-03-21T13:12:33Z)
• Attention-based Feature Compression for CNN Inference Offloading in Edge Computing [93.67044879636093]
  This paper studies the computational offloading of CNN inference in device-edge co-inference systems. We propose a novel autoencoder-based CNN architecture (AECNN) for effective feature extraction at end-device. Experiments show that AECNN can compress the intermediate data by more than 256x with only about 4% accuracy loss.
  arXiv  Detail & Related papers  (2022-11-24T18:10:01Z)
• DeltaCNN: End-to-End CNN Inference of Sparse Frame Differences in Videos [16.644938608211202]
  Convolutional neural network inference on video data requires powerful hardware for real-time processing. We present a sparse convolutional neural network framework that enables sparse frame-by-frame updates. We are the first to significantly outperform the dense reference, cuDNN, in practical settings, achieving speedups of up to 7x with only marginal differences in accuracy.
  arXiv  Detail & Related papers  (2022-03-08T10:54:00Z)
• Implicit Neural Video Compression [17.873088127087605]
  We propose a method to compress full-resolution video sequences with implicit neural representations. Each frame is represented as a neural network that maps coordinate positions to pixel values. We use a separate implicit network to modulate the coordinate inputs, which enables efficient motion compensation between frames.
  arXiv  Detail & Related papers  (2021-12-21T15:59:00Z)
• MoViNets: Mobile Video Networks for Efficient Video Recognition [52.49314494202433]
  3D convolutional neural networks (CNNs) are accurate at video recognition but require large computation and memory budgets. We propose a three-step approach to improve computational efficiency while substantially reducing the peak memory usage of 3D CNNs.
  arXiv  Detail & Related papers  (2021-03-21T23:06:38Z)
• Fixed-point Quantization of Convolutional Neural Networks for Quantized Inference on Embedded Platforms [0.9954382983583577]
  We propose a method to optimally quantize the weights, biases and activations of each layer of a pre-trained CNN. We find that layer-wise quantization of parameters significantly helps in this process.
  arXiv  Detail & Related papers  (2021-02-03T17:05:55Z)
• The Mind's Eye: Visualizing Class-Agnostic Features of CNNs [92.39082696657874]
  We propose an approach to visually interpret CNN features given a set of images by creating corresponding images that depict the most informative features of a specific layer. Our method uses a dual-objective activation and distance loss, without requiring a generator network nor modifications to the original model.
  arXiv  Detail & Related papers  (2021-01-29T07:46:39Z)
• L2PF -- Learning to Prune Faster [57.32153461504626]
  We present a multi-task, try-and-learn method, discretely learning redundant filters of the CNN and a continuous action of how long the layers have to be fine-tuned. For ResNet20, we have achieved a compression ratio of 3.84 x with minimal accuracy degradation. Compared to the state-of-the-art pruning method, we reduced the GPU hours by 1.71 x.
  arXiv  Detail & Related papers  (2021-01-07T18:13:37Z)
• Efficient Integer-Arithmetic-Only Convolutional Neural Networks [87.01739569518513]
  We replace conventional ReLU with Bounded ReLU and find that the decline is due to activation quantization. Our integer networks achieve equivalent performance as the corresponding FPN networks, but have only 1/4 memory cost and run 2x faster on modern GPU.
  arXiv  Detail & Related papers  (2020-06-21T08:23:03Z)
• Approximation and Non-parametric Estimation of ResNet-type Convolutional Neural Networks [52.972605601174955]
  We show a ResNet-type CNN can attain the minimax optimal error rates in important function classes. We derive approximation and estimation error rates of the aformentioned type of CNNs for the Barron and H"older classes.
  arXiv  Detail & Related papers  (2019-03-24T19:42:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.