# HS-ResNet: Hierarchical-Split Block on Convolutional Neural Network

arXiv: https://arxiv.org/abs/2010.07621 · HTML: https://mlsanity.com/paper/2010.07621/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Oct 2020.

## Built on by (most cited first, 6 of 6)

- [EPSANet: An Efficient Pyramid Squeeze Attention Block on Convolutional Neural Network](https://mlsanity.com/paper/2105.14447/) (2021)
- [GhostSR: Learning Ghost Features for Efficient Image Super-Resolution](https://mlsanity.com/paper/2101.08525/) (2021)
- [V2L: Leveraging Vision and Vision-language Models into Large-scale Product Retrieval](https://mlsanity.com/paper/2207.12994/) (2022)
- [TMS: A Temporal Multi-scale Backbone Design for Speaker Embedding](https://mlsanity.com/paper/2203.09098/) (2022)
- [Multi-scale Network with Attentional Multi-resolution Fusion for Point Cloud Semantic Segmentation](https://mlsanity.com/paper/2206.13628/) (2022)
- [Multi-view learning for automatic classification of multi-wavelength auroral images](https://mlsanity.com/paper/2311.02947/) (2023)

## Abstract (arXiv)

This paper addresses representational block named Hierarchical-Split Block, which can be taken as a plug-and-play block to upgrade existing convolutional neural networks, improves model performance significantly in a network. Hierarchical-Split Block contains many hierarchical split and concatenate connections within one single residual block. We find multi-scale features is of great importance for numerous vision tasks. Moreover, Hierarchical-Split block is very flexible and efficient, which provides a large space of potential network architectures for different applications. In this work, we present a common backbone based on Hierarchical-Split block for tasks: image classification, object detection, instance segmentation and semantic image segmentation/parsing. Our approach shows significant improvements over all these core tasks in comparison with the baseline. As shown in Figure1, for image classification, our 50-layers network(HS-ResNet50) achieves 81.28% top-1 accuracy with competitive latency on ImageNet-1k dataset. It also outperforms most state-of-the-art models. The source code and models will be available on: https://github.com/PaddlePaddle/PaddleClas