arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2302.01384cs.CV
收藏

视觉模型的能量激励自监督预训练

Energy-Inspired Self-Supervised Pretraining for Vision Models

Purdue University(普渡大学);Microsoft Corporation(微软公司)

Ze Wang, Jiang Wang, Zicheng Liu, Qiang Qiu

摘要:

基于深度网络的前向和后向传递自然形成输入和输出表示之间的对称映射这一事实,我们引入了一个简单而有效的自监督视觉模型预训练框架,该框架受到基于能量的模型(EBM)的启发。在所提出的框架中,我们将能量估计和数据恢复建模为单个网络的前向和后向传递,而没有任何辅助组件,例如,额外的解码器对于前向传递,我们将网络拟合到一个能量函数,该函数将低能量分数分配给属于未标记数据集的样本,否则将高能量。对于向后传递,我们使用基于梯度的优化沿着能量最小化的方向迭代地从损坏的版本恢复数据。通过这种方式,我们自然地将广泛用于掩码图像建模的编码器-解码器架构折叠到单个视觉模型的前向和后向通道中。因此,我们的框架现在接受各种各样的借口任务与不同的数据损坏方法,并允许模型从掩码图像建模,补丁排序和图像复原,包括超分辨率,去噪和着色进行预训练。我们通过大量的实验来支持我们的研究结果,并表明与最先进的自监督视觉模型预训练方法相比,所提出的方法可以提供相当甚至更好的性能,并且训练时间明显更少。我们的研究结果为进一步探索自监督视觉模型预训练和掩码图像建模之外的任务提供了启发。

英文摘要:

Motivated by the fact that forward and backward passes of a deep network naturally form symmetric mappings between input and output representations, we introduce a simple yet effective self-supervised vision model pretraining framework inspired by energy-based models (EBMs). In the proposed framework, we model energy estimation and data restoration as the forward and backward passes of a single network without any auxiliary components, e.g., an extra decoder. For the forward pass, we fit a network to an energy function that assigns low energy scores to samples that belong to an unlabeled dataset, and high energy otherwise. For the backward pass, we restore data from corrupted versions iteratively using gradient-based optimization along the direction of energy minimization. In this way, we naturally fold the encoder-decoder architecture widely used in masked image modeling into the forward and backward passes of a single vision model. Thus, our framework now accepts a wide range of pretext tasks with different data corruption methods, and permits models to be pretrained from masked image modeling, patch sorting, and image restoration, including super-resolution, denoising, and colorization. We support our findings with extensive experiments, and show the proposed method delivers comparable and even better performance with remarkably fewer epochs of training compared to the state-of-the-art self-supervised vision model pretraining methods. Our findings shed light on further exploring self-supervised vision model pretraining and pretext tasks beyond masked image modeling.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑