Diverse Branch Block: Building a Convolution as an Inception-like Unit

By Xiaohan Ding, X. Zhang, Jungong Han, Guiguang Ding. Computer Vision and Pattern Recognition, 2021

436 citations recorded by the source corpus.

We propose a universal building block of Convolutional Neural Network (ConvNet) to improve the performance without any inference-time costs. The block is named Diverse Branch Block (DBB), which enhances the representational capacity of a single convolution by combining diverse branches of different scales and complexities to enrich the feature space, including sequences of convolutions, multiscale convolutions, and average pooling. After training, a DBB can be equivalently converted into a single conv layer for deployment. Unlike the advancements of novel ConvNet architectures, DBB complicates the training-time microstructure while maintaining the macro architecture, so that it can be used as a drop-in replacement for regular conv layers of any architecture. In this way, the model can be trained to reach a higher level of performance and then transformed into the original inference-time structure for inference. DBB improves ConvNets on image classification (up to 1.9% higher top-1 accuracy on ImageNet), object detection and semantic segmentation. The PyTorch code and models are released at https://github.com/DingXiaoH/DiverseBranchBlock.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/1f34aee634327c6d4f13f9018c34df4d3239ecb6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors (2022, methodology): “Model re-parametrization techniques [71, 31, 75, 19, 33, 11, 4, 24, 13, 12, 10, 29, 14, 78] merge multiple computational modules into one at inference stage.”
FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization (2023, methodology): “In order to increase capacity of the these layers, we use linear train-time overparameterization as introduced in [13, 11, 12, 57, 18].”
MobileOne: An Improved One millisecond Mobile Backbone (2022, background): “Expand-Nets [19], ACNet [11] and DBBNet [12], propose a drop-in replacement for a regular convolution layer in recent CNN architectures and show improvements in accuracy.”