Efficient Facial Landmark Detection for Embedded Systems

This paper introduces the Efficient Facial Landmark Detection (EFLD) model, specifically designed for edge devices confronted with the challenges related to power consumption and time latency. EFLD features a lightweight backbone and a flexible detection head, each significantly enhancing operational efficiency on resource-constrained devices. To improve the model's robustness, we propose a cross-format training strategy. This strategy leverages a wide variety of publicly accessible datasets to enhance the model's generalizability and robustness, without increasing inference costs. Our ablation study highlights the significant impact of each component on reducing computational demands, model size, and improving accuracy. EFLD demonstrates superior performance compared to competitors in the IEEE ICME 2024 Grand Challenges PAIR Competition, a contest focused on low-power, efficient, and accurate facial-landmark detection for embedded systems, showcasing its effectiveness in real-world facial landmark detection tasks.

Image-Text Co-Decomposition for Text-Supervised Semantic Segmentation

This paper addresses text-supervised semantic segmentation, aiming to learn a model capable of segmenting arbitrary visual concepts within images by using only image-text pairs without dense annotations. Existing methods have demonstrated that contrastive learning on image-text pairs effectively aligns visual segments with the meanings of texts. We notice that there is a discrepancy between text alignment and semantic segmentation: A text often consists of multiple semantic concepts, whereas semantic segmentation strives to create semantically homogeneous segments. To address this issue, we propose a novel framework, Image-Text Co-Decomposition (CoDe), where the paired image and text are jointly decomposed into a set of image regions and a set of word segments, respectively, and contrastive learning is developed to enforce region-word alignment. To work with a vision-language model, we present a prompt learning mechanism that derives an extra representation to highlight an image segment or a word segment of interest, with which more effective features can be extracted from that segment. Comprehensive experimental results demonstrate that our method performs favorably against existing text-supervised semantic segmentation methods on six benchmark datasets.