Partial Channel Network: Compute Fewer, Perform Better

• URL: http://arxiv.org/abs/2502.01303v1
• Date: Mon, 03 Feb 2025 12:26:55 GMT
• Title: Partial Channel Network: Compute Fewer, Perform Better
• Authors: Haiduo Huang, Tian Xia, Wenzhe zhao, Pengju Ren,
• Abstract summary: We propose a new partial channel mechanism (PCM) to exploit redundancy within feature map channels. We introduce a novel partial attention convolution (PATConv) that can efficiently combine convolution with visual attention. Building on PATConv and DPConv, we propose a new hybrid network family, named PartialNet, which achieves superior top-1 accuracy and inference speed.
• Score: 6.666628122653455
• License:
• Abstract: Designing a module or mechanism that enables a network to maintain low parameters and FLOPs without sacrificing accuracy and throughput remains a challenge. To address this challenge and exploit the redundancy within feature map channels, we propose a new solution: partial channel mechanism (PCM). Specifically, through the split operation, the feature map channels are divided into different parts, with each part corresponding to different operations, such as convolution, attention, pooling, and identity mapping. Based on this assumption, we introduce a novel partial attention convolution (PATConv) that can efficiently combine convolution with visual attention. Our exploration indicates that the PATConv can completely replace both the regular convolution and the regular visual attention while reducing model parameters and FLOPs. Moreover, PATConv can derive three new types of blocks: Partial Channel-Attention block (PAT_ch), Partial Spatial-Attention block (PAT_sp), and Partial Self-Attention block (PAT_sf). In addition, we propose a novel dynamic partial convolution (DPConv) that can adaptively learn the proportion of split channels in different layers to achieve better trade-offs. Building on PATConv and DPConv, we propose a new hybrid network family, named PartialNet, which achieves superior top-1 accuracy and inference speed compared to some SOTA models on ImageNet-1K classification and excels in both detection and segmentation on the COCO dataset. Our code is available at https://github.com/haiduo/PartialNet.

Related papers

• SCHEME: Scalable Channel Mixer for Vision Transformers [52.605868919281086]
  Vision Transformers have achieved impressive performance in many vision tasks. Much less research has been devoted to the channel mixer or feature mixing block (FFN or) We show that the dense connections can be replaced with a diagonal block structure that supports larger expansion ratios.
  arXiv  Detail & Related papers  (2023-12-01T08:22:34Z)
• Bridging the Gap Between Vision Transformers and Convolutional Neural Networks on Small Datasets [91.25055890980084]
  There still remains an extreme performance gap between Vision Transformers (ViTs) and Convolutional Neural Networks (CNNs) when training from scratch on small datasets. We propose Dynamic Hybrid Vision Transformer (DHVT) as the solution to enhance the two inductive biases. Our DHVT achieves a series of state-of-the-art performance with a lightweight model, 85.68% on CIFAR-100 with 22.8M parameters, 82.3% on ImageNet-1K with 24.0M parameters.
  arXiv  Detail & Related papers  (2022-10-12T06:54:39Z)
• Group Fisher Pruning for Practical Network Compression [58.25776612812883]
  We present a general channel pruning approach that can be applied to various complicated structures. We derive a unified metric based on Fisher information to evaluate the importance of a single channel and coupled channels. Our method can be used to prune any structures including those with coupled channels.
  arXiv  Detail & Related papers  (2021-08-02T08:21:44Z)
• CT-Net: Channel Tensorization Network for Video Classification [48.4482794950675]
  3D convolution is powerful for video classification but often computationally expensive. Most approaches fail to achieve a preferable balance between convolutional efficiency and feature-interaction sufficiency. We propose a concise and novel Channelization Network (CT-Net) Our CT-Net outperforms a number of recent SOTA approaches, in terms of accuracy and/or efficiency.
  arXiv  Detail & Related papers  (2021-06-03T05:35:43Z)
• Skip-Convolutions for Efficient Video Processing [21.823332885657784]
  Skip-Convolutions leverage the large amount of redundancies in video streams and save computations. We replace all convolutions with Skip-Convolutions in two state-of-the-art architectures, namely EfficientDet and HRNet. We reduce their computational cost consistently by a factor of 34x for two different tasks, without any accuracy drop.
  arXiv  Detail & Related papers  (2021-04-23T09:10:39Z)
• Diverse Branch Block: Building a Convolution as an Inception-like Unit [123.59890802196797]
  We propose a universal building block of Convolutional Neural Network (ConvNet) to improve the performance without any inference-time costs. The Diverse Branch Block (DBB) enhances the representational capacity of a single convolution by combining diverse branches of different scales and complexities. After training, a DBB can be equivalently converted into a single conv layer for deployment.
  arXiv  Detail & Related papers  (2021-03-24T18:12:00Z)
• PSConv: Squeezing Feature Pyramid into One Compact Poly-Scale Convolutional Layer [76.44375136492827]
  Convolutional Neural Networks (CNNs) are often scale-sensitive. We bridge this regret by exploiting multi-scale features in a finer granularity. The proposed convolution operation, named Poly-Scale Convolution (PSConv), mixes up a spectrum of dilation rates.
  arXiv  Detail & Related papers  (2020-07-13T05:14:11Z)
• Split to Be Slim: An Overlooked Redundancy in Vanilla Convolution [11.674837640798126]
  We propose a textbfsplit based textbfconvolutional operation, namely SPConv, to tolerate features with similar patterns but require less computation. We show that SPConv-equipped networks consistently outperform state-of-the-art baselines in both accuracy and inference time on GPU.
  arXiv  Detail & Related papers  (2020-06-22T09:08:51Z)
• Dynamic Region-Aware Convolution [85.20099799084026]
  We propose a new convolution called Dynamic Region-Aware Convolution (DRConv), which can automatically assign multiple filters to corresponding spatial regions. On ImageNet classification, DRConv-based ShuffleNetV2-0.5x achieves state-of-the-art performance of 67.1% at 46M multiply-adds level with 6.3% relative improvement.
  arXiv  Detail & Related papers  (2020-03-27T05:49:57Z)
• SlimConv: Reducing Channel Redundancy in Convolutional Neural Networks by Weights Flipping [43.37989928043927]
  We design a novel Slim Convolution (SlimConv) module to boost the performance of CNNs by reducing channel redundancies. SlimConv consists of three main steps: Reconstruct, Transform and Fuse, through which the features are splitted and reorganized in a more efficient way. We validate the effectiveness of SlimConv by conducting comprehensive experiments on ImageNet, MS2014, Pascal VOC2012 segmentation, and Pascal VOC2007 detection datasets.
  arXiv  Detail & Related papers  (2020-03-16T23:23:10Z)

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.