Online Convolutional Reparameterization

• Published in: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) pp. 558 - 567
• Main Authors: Hu, Mu, Feng, Junyi, Hua, Jiashen, Lai, Baisheng, Huang, Jianqiang, Gong, Xiaojin, Hua, Xiansheng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2022
• Subjects: Biological system modeling; Computational modeling; Computer vision; Convolution; Convolutional codes; Costs; Deep learning architectures and techniques; Efficient learning and inferences; Training
• ISSN: 2575-7075
• DOI: 10.1109/CVPR52688.2022.00065

Structural re-parameterization has drawn increasing attention in various computer vision tasks. It aims at improving the performance of deep models without introducing any inference-time cost. Though efficient during inference, such models rely heavily on the complicated training-time blocks to achieve high accuracy, leading to large extra training cost. In this paper, we present online convolutional re-parameterization (OREPA), a two-stage pipeline, aiming to reduce the huge training overhead by squeezing the complex training-time block into a single convolution. To achieve this goal, we introduce a linear scaling layer for better optimizing the online blocks. Assisted with the reduced training cost, we also explore some more effective re-param components. Compared with the state-of-the-art re-param models, OREPA is able to save the training-time memory cost by about 70% and accelerate the training speed by around 2×. Meanwhile, equipped with OREPA, the models out-perform previous methods on ImageNet by up to +0.6%. We also conduct experiments on object detection and semantic segmentation and show consistent improvements on the downstream tasks. Codes are available at https://github.com/JUGGHM/OREPA_CVPR2022.