Multimodal Markov Random Field for Image Reranking Based on Relevance Feedback
Figure 8
Comparison of the multimodal MRF (blue squares) with the base image retrieval system (green upwards-triangles), the textual MRF (red rhombus), and the visual MRF (yellow downwards-triangles). The plot shows the average MAP of each method (y-axis) against the number of relevant feedback images considered (x-axis).
