2018/05/28 by François Plesse, Alexandru Ginsca, Plesse, François +5
Biochemistry, Genetics and Molecular Biology · Computer Science · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Genomics and Phylogenetic Studies #Image Retrieval and Classification Techniques #cs.CV
paper · pdf · doi:10.48550/arxiv.1805.10802
Accepted submission to ICME 2018
arxiv created 2018/05/28 · openalex publication_date 2018/05/28 · arxiv updated 2018/05/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
A thorough comprehension of image content demands a complex grasp of the interactions that may occur in the natural world. One of the key issues is to describe the visual relationships between objects. When dealing with real world data, capturing these very diverse interactions is a difficult problem. It can be alleviated by incorporating common sense in a network. For this, we propose a framework that makes use of semantic knowledge and estimates the relevance of object pairs during both training and test phases. Extracted from precomputed models and training annotations, this information is distilled into the neural network dedicated to this task. Using this approach, we observe a significant improvement on all classes of Visual Genome, a challenging visual relationship dataset. A 68.5% relative gain on the recall at 100 is directly related to the relevance estimate and a 32.7% gain to the knowledge distillation.