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 2105.05409cs.CVcs.MM
收藏

一种大规模食品图像分割基准算法

A Large-Scale Benchmark for Food Image Segmentation

Singapore Management University(新加坡管理大学);Beijing Jiaotong University(北京交通大学);Salesforce Research Asia(亚洲 Salesforce 研究院)

Xiongwei Wu, Xin Fu, Ying Liu, Ee-Peng Lim, Steven C. H. Hoi, Qianru Sun

摘要:

食品图像分割是开发与健康相关的应用(如估计食品热量和营养素)的关键和不可缺少的任务。现有的食物图像分割模型由于两个原因而表现不佳:(1)缺乏具有细粒度成分标签和逐像素位置掩模的高质量食物图像数据集-现有的数据集携带粗糙的成分标签或者尺寸小;以及(2)食物的复杂外观使得难以定位和识别食物图像中的成分,例如,在相同的图像中,配料可以彼此重叠,并且相同的配料可以在不同的食物图像中明显地出现。在这项工作中,我们构建了一个新的食物图像数据集FoodSeg 103(及其扩展FoodSeg 154),包含9,490张图像。我们用154个成分类来注释这些图像,每个图像平均有6个成分标签和像素掩码。此外,我们提出了一种多模态预训练方法,称为ReLeM,它明确地为分割模型配备了丰富的语义食物知识。在实验中,我们使用三种流行的语义分割方法(即,基于扩张卷积、基于特征金字塔和基于视觉Transformer)作为基线,并在我们的新数据集上评估它们以及ReLeM。我们相信FoodSeg 103(及其扩展FoodSeg 154)和使用ReLeM的预训练模型可以作为基准,促进未来对细粒度食物图像理解的工作。我们在\url{https://xiongweiwu.github.io/foodseg103.html}上公开了所有这些数据集和方法。

英文摘要:

Food image segmentation is a critical and indispensible task for developing health-related applications such as estimating food calories and nutrients. Existing food image segmentation models are underperforming due to two reasons: (1) there is a lack of high quality food image datasets with fine-grained ingredient labels and pixel-wise location masks -- the existing datasets either carry coarse ingredient labels or are small in size; and (2) the complex appearance of food makes it difficult to localize and recognize ingredients in food images, e.g., the ingredients may overlap one another in the same image, and the identical ingredient may appear distinctly in different food images. In this work, we build a new food image dataset FoodSeg103 (and its extension FoodSeg154) containing 9,490 images. We annotate these images with 154 ingredient classes and each image has an average of 6 ingredient labels and pixel-wise masks. In addition, we propose a multi-modality pre-training approach called ReLeM that explicitly equips a segmentation model with rich and semantic food knowledge. In experiments, we use three popular semantic segmentation methods (i.e., Dilated Convolution based, Feature Pyramid based, and Vision Transformer based) as baselines, and evaluate them as well as ReLeM on our new datasets. We believe that the FoodSeg103 (and its extension FoodSeg154) and the pre-trained models using ReLeM can serve as a benchmark to facilitate future works on fine-grained food image understanding. We make all these datasets and methods public at \url{https://xiongweiwu.github.io/foodseg103.html}.

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。

↑