# LowFormer: Hardware Efficient Design for Convolutional Transformer Backbones

Source: [LowFormer: Hardware Efficient Design for Convolutional Transformer Backbones](<https://www.sota2.com/research/paper/68fa74c21e209131cea639a5>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:57:09.300Z

Authors: Moritz Nottebaum, Matteo Dunnhofer, Christian Micheloni

Published: 2024-09-05T12:18:32.000Z

## Abstract

Research in efficient vision backbones is evolving into models that are a mixture of convolutions and transformer blocks. A smart combination of both, architecture-wise and component-wise is mandatory to excel in the speedaccuracy trade-off. Most publications focus on maximizing accuracy and utilize MACs (multiply accumulate operations) as an efficiency metric. The latter however often do not measure accurately how fast a model actually is due to factors like memory access cost and degree of parallelism. We analyzed common modules and architectural design choices for backbones not in terms of MACs, but rather in actual throughput and latency, as the combination of the latter two is a better representation of the efficiency of models in real applications. We applied the conclusions taken from that analysis to create a recipe for increasing hardware-efficiency in macro design. Additionally we introduce a simple slimmed-down version of MultiHead Self-Attention, that aligns with our analysis. We combine both macro and micro design to create a new family of hardware-efficient backbone networks called LowFormer. LowFormer achieves a remarkable speedup in terms of throughput and latency, while achieving similar or better accuracy than current state-of-the-art efficient backbones. In order to prove the generalizability of our hardware-efficient design, we evaluate our method on GPU, mobile GPU and ARM CPU. We further show that the downstream tasks object detection and semantic segmentation profit from our hardware-efficient architecture. Code and models are available at https://github.com/ altair199797/LowFormer.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Object Detection on COCO (mAP)](<https://www.sota2.com/research/sota/object-detection-on-coco-map.md>): Object Detection; COCO; mAP: 41.4.

- [Image Classification on ImageNet-1K 1.0 (val)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-1-0-val.md>): Image Classification; ImageNet-1K 1.0 (val); Top-1 Accuracy: 83.6.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2409.03460>)

[PDF](<https://arxiv.org/pdf/2409.03460.pdf>)
