VIPL's 1 paper is accepted by ACM MM 2023

Congratulations! VIPL's 1 paper is accepted by ACM MM (ACM International Conference on Multimedia) 2023! ACM MM is a top-tier international conference on multimedia. This year, ACM MM will be held in Ottawa Canada on Oct the 29th through Nov the 3rd.

The accepted papers are summarized as follows:

1. SeeDS: Semantic Separable Diffusion Synthesizer for Zero-Shot Food Detection(Pengfei Zhou, Weiqing Min, Yang Zhang, Jiajun Song, Ying Jin, and Shuqiang Jiang)

Food detection is a fundamental task in food computing that supports various multimedia applications, including food recommendation and dietary monitoring. To deal with real-world scenarios, food detection needs to localize and recognize novel food objects that are not seen during training, which requires Zero-Shot Detection (ZSD). However, it is challenging for current ZSD methods to distinguish fine-grained food classes due to the complexity of semantic attributes and the diversity of intra-class features. To address these issues, we propose a novel Zero-Shot Food Detection (ZSFD) framework named Semantic Separable Diffusion Synthesizer (SeeDS). SeeDS consists of two modules: a Semantic Separable Synthesizing Module (M) and a Region Feature Denoising Diffusion Model (RFDDM). The designed M learns the disentangled semantic representation for complex food attributes from ingredients and cuisines, and synthesizes discriminative food features via enhanced semantic information. The proposed RFDDM utilizes a novel diffusion model to generate more diversified region features and enhances ZSFD via fine-grained synthesized features. Extensive experiments show the state-of-the-art ZSFD performance of our proposed method on two food datasets, ZSFooD and UECFOOD-256. We also apply our method to two general ZSD datasets, PASCAL VOC and MS COCO, and demonstrate its effectiveness on widely-used ZSD datasets.