Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Overview of the proposed method.

Example query from the WikiArt dataset, where to is initial IR results, and tn is Rococo’ conditioned IR results. The proposed method updates embedding vectors to to tn through ti by exploring the embedding space. A green box indicates that the retrieved image has the same class as the condition.

More »

Fig 1 Expand

Fig 2.

Proposed conditional image retrieval method.

More »

Fig 2 Expand

Fig 3.

Proposed Backward Search algorithm for conditional image retrieval.

More »

Fig 3 Expand

Fig 4.

Proposed knowledge distillation model for the Backward Search.

More »

Fig 4 Expand

Table 1.

Datasets and queries used in the experiments.

More »

Table 1 Expand

Table 2.

Results of the Backward Search performance without regularization.

More »

Table 2 Expand

Fig 5.

Backward Search results using various λ regularization parameter settings.

(A) t-SNE scatterplot depicting the embedding space of the Baroque and Impressionism dataset embeddings from WikiArt. The ‘init’ query image is marked on the plot. (B) Conditional image retrieval (CIR) result for the query image in (A). The embedding vectors of the top three retrieved images are averaged and visualized on the same t-SNE scatterplot in (B). A green box indicates that the retrieved image has the same class as the condition.

More »

Fig 5 Expand

Fig 6.

Variation in performance of mAP@10 and mTS@10 based on the lambda value of the regularization term in the test dataset.

More »

Fig 6 Expand

Table 3.

Experiments involving variations in model architectures and backbone fine-tuning.

More »

Table 3 Expand

Table 4.

Comparison study between the proposed method and state-of-the-art CoIR models on the CUB, aPY, and WikiArt datasets.

More »

Table 4 Expand

Fig 7.

Qualitative result of WikiArt test dataset with WikiArt-style query.

We mark correct retrieved images with green boxes for the best view.

More »

Fig 7 Expand

Fig 8.

Qualitative result of WikiArt test dataset with WikiArt-multi query.

We mark correct retrieved images with green boxes for the best view.

More »

Fig 8 Expand

Table 5.

Iterations and time consumption per query for the Backward Search.

More »

Table 5 Expand

Table 6.

Results of knowledge distillation of the Backward Search.

More »

Table 6 Expand

Fig 9.

Qualitative multiple condition image retrieval results.

We marked the labels of the WikiArt dataset on each retrieved image with colored circles. purple is "sketch and study", green is "landscape", and blue is "Post Impressionism".

More »

Fig 9 Expand