Image-based dietary assessment serves as an efficient and accurate solution for recording and analyzing nutrition intake using eating occasion images as input. Deep learning-based techniques are commonly used to perform image analysis such as food classification, segmentation, and portion size estimation, which rely on large amounts of food images with annotations for training. However, such data dependency poses significant barriers to real-world applications, because acquiring a substantial, diverse, and balanced set of food images can be challenging. One potential solution is to use synthetic food images for data augmentation. Although existing work has explored the use of generative adversarial networks (GAN) based structures for generation, the quality of synthetic food images still remains subpar. In addition, while diffusion-based generative models have shown promising results for general image generation tasks, the generation of food images can be challenging due to the substantial intra-class variance. In this paper, we investigate the generation of synthetic food images based on the conditional diffusion model and propose an effective clustering-based training framework, named ClusDiff, for generating high-quality and representative food images. The proposed method is evaluated on the Food-101 dataset and shows improved performance when compared with existing image generation works. We also demonstrate that the synthetic food images generated by ClusDiff can help address the severe class imbalance issue in long-tailed food classification using the VFN-LT dataset.
5 September 2022 in IEEE Geoscience and Remote Sensing Letters
Detection of small, point targets is fundamental in applications, such as early warning systems, surveillance, astronomy, and microscopy. The presence of noise and clutter can make it challenging to detect small targets while minimizing false detections. This letter presents a method for infrared small target detection using convolutional neural networks (CNNs). The proposed method augments a conventional space-based detection processing chain with a lightweight neural network to predict the probability that a detection is a target. The proposed network is trained on $7 \times 7$ pixel windows using both the image sequence and the respective background-subtracted images. Results show that our method improves the probability of detection at low false detection rates.