Donghwan Lee,
Donghwan Lee
Institution:
Email:
Wooju Kim
Wooju Kim
Institution:
Email:
Conditional image retrieval (CIR), which involves retrieving images by a query image along with user-specified conditions, is essential in computer vision research for efficient image search and automated image analysis. The existing approaches, such as composed image retrieval (CoIR) methods, have ...
More
Conditional image retrieval (CIR), which involves retrieving images by a query image along with user-specified conditions, is essential in computer vision research for efficient image search and automated image analysis. The existing approaches, such as composed image retrieval (CoIR) methods, have been actively studied. However, these methods face challenges as they require either a triplet dataset or richly annotated image-text pairs, which are expensive to obtain. In this work, we demonstrate that CIR at the image-level concept can be achieved using an inverse mapping approach that explores the model’s inductive knowledge. Our proposed CIR method, called Backward Search, updates the query embedding to conform to the condition. Specifically, the embedding of the query image is updated by predicting the probability of the label and minimizing the difference from the condition label. This enables CIR with image-level concepts while preserving the context of the query. In this paper, we introduce the Backward Search method that enables single and multi-conditional image retrieval. Moreover, we efficiently reduce the computation time by distilling the knowledge. We conduct experiments using the WikiArt, aPY, and CUB benchmark datasets. The proposed method achieves an average mAP@10 of 0.541 on the datasets, demonstrating a marked improvement compared to the CoIR methods in our comparative experiments. Furthermore, by employing knowledge distillation with the Backward Search model as the teacher, the student model achieves a significant reduction in computation time, up to 160 times faster with only a slight decrease in performance. The implementation of our method is available at the following URL: https://github.com/dhlee-work/BackwardSearch.
Less
2 days ago
Slawomira Hajduk
Slawomira Hajduk
Institution: Faculty of Engineering Management, Bialystok University of Technology,
Email: s.hajduk@pb.edu.pl
The effects of urban transport are highly concerning. The rapid urbanization and motorization in smart cities have a huge impact on sustainability. The goal of the paper is to analyse the smart
cities selected, in terms of the urban transport. This paper presents an overview of research works
pub...
More
The effects of urban transport are highly concerning. The rapid urbanization and motorization in smart cities have a huge impact on sustainability. The goal of the paper is to analyse the smart
cities selected, in terms of the urban transport. This paper presents an overview of research works
published between 1991 and 2020 concerning urban transport and MCDM (multi-criteria decision
making). The author highlights the importance of decision-making criteria and their weight, as well as
techniques. Seven criteria and forty-four objects were used as the input of the approach. The entropy
weight method was used to compute the weight of each criterion. The TOPSIS (Technique for Order
Performance by Similarity to Ideal Solution) was applied to calculate the assessment and ranking of
transport performance for each smart city. Portland was found to be the best location for transport
enterprises and projects; Tbilisi was ranked last. The values of the relative closeness coefficient ranged
from 0.03504 to 0.921402. Finally, some suggestions for future research are discussed.
Less
2 years ago