Hyperbolic Image Embeddings
Computer vision tasks such as image classification, image retrieval and few-shot learning are currently dominated by Euclidean and spherical embeddings, so that the final decisions about class belongings or the degree of similarity are made using linear hyperplanes, Euclidean distances, or spherical geodesic distances (cosine similarity). In this work, we demonstrate that in many practical scenarios hyperbolic embeddings provide a better alternative.
Code (3)
Tasks
Few-Shot LearningGeneral Classificationimage-classificationImage ClassificationImage RetrievalRetrievalSimilar Papers 제목 키워드 기반
Hyperbolic vs Euclidean Embeddings in Few-Shot Learning: Two Sides of the Same Coin
Recent research in representation learning has shown that hierarchical data lends itself to low-dimensional and highly informative representations in hyperbolic space. However, even if hyperbolic embeddings have gathered…
Few-Shot LearningRepresentation LearningUnderstanding Fine-tuning CLIP for Open-vocabulary Semantic Segmentation in Hyperbolic Space
CLIP, a foundational vision-language model, has emerged as a powerful tool for open-vocabulary semantic segmentation. While freezing the text encoder preserves its powerful embeddings, recent studies show that fine-t…
Open Vocabulary Semantic SegmentationOpen-Vocabulary Semantic SegmentationSegmentationSemantic SegmentationHyperbolic Learning with Multimodal Large Language Models
Hyperbolic embeddings have demonstrated their effectiveness in capturing measures of uncertainty and hierarchical relationships across various deep-learning tasks, including image segmentation and active learning. Howeve…
Active LearningImage SegmentationSemantic SegmentationHyperbolic Image Segmentation
For image segmentation, the current standard is to perform pixel-level optimization and inference in Euclidean output embedding spaces through linear hyperplanes. In this work, we show that hyperbolic manifolds provide a…
Image SegmentationSegmentationSemantic SegmentationSkip-gram word embeddings in hyperbolic space
Recent work has demonstrated that embeddings of tree-like graphs in hyperbolic space surpass their Euclidean counterparts in performance by a large margin. Inspired by these results and scale-free structure in the word c…
Learning Word EmbeddingsWord EmbeddingsWord Similarity