Existing facial expression recognition (FER) methods are mainly devoted to learning discriminative features from normal-light images. However, their performance drops sharply when they are used for low-light images. In this paper, we propose a novel low-light FER framework (termed LL-FER) that can simultaneously enhance the images and recognition tasks of low-light facial expression images. Specifically, we first meticulously design a low-light enhancement network (LLENet) to recover expressions images’ rich detail information. Then, we design a joint loss to train the LLENet with FER network in a cascade manner, so that the FER network can guide the LLENet to gradually perceive and restore discriminative features which are useful for FER during the training process. Extensive experiments show that the LLENet not only achieves competitive results both quantitatively and qualitatively, but also in the LL-FER framework, which can produce results more suitable for FER tasks, further improving the performance of the FER methods.

Jie Ou, Yueming Chen, Prof. Tian. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 6: Industry Track). 2024.

Current state-of-the-art medical image segmentation methods prioritize accuracy but often at the expense of increased computational demands and larger model sizes. Applying these large-scale models to the relatively limited scale of medical image datasets tends to induce redundant computation, complicating the process without the necessary benefits. This approach not only adds complexity but also presents challenges for the integration and deployment of lightweight models on edge devices. For instance, recent transformer-based models have excelled in 2D and 3D medical image segmentation due to their extensive receptive fields and high parameter count. However, their effectiveness comes with a risk of overfitting when applied to small datasets and often neglects the vital inductive biases of Convolutional Neural Networks (CNNs), essential for local feature representation. In this work, we propose PMFSNet, a novel medical imaging segmentation model that effectively balances global and local feature processing while avoiding the computational redundancy typical in larger models. PMFSNet streamlines the UNet-based hierarchical structure and simplifies the self-attention mechanism's computational complexity, making it suitable for lightweight applications. It incorporates a plug-and-play PMFS block, a multi-scale feature enhancement module based on attention mechanisms, to capture long-term dependencies. Extensive comprehensive results demonstrate that even with a model (less than 1 million parameters), our method achieves superior performance in various segmentation tasks across different data scales. It achieves (IoU) metrics of 84.68%, 82.02%, and 78.82% on public datasets of teeth CT (CBCT), ovarian tumors ultrasound(MMOTU), and skin lesions dermoscopy images (ISIC 2018), respectively. The source code is available at https://github.com/yykzjh/PMFSNet.