Unlock the power of visual search with Image Retrieval PGVector! Learn to build a system using PostgreSQL, MinIO, & image embeddings. Check it out! #ImageRetrieval #PGVector #VisualSearch
https://teguhteja.id/image-retrieval-pgvector-guide/?utm_source=mastodon&utm_medium=jetpack_social
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.
COMPUTER VISION / IMAGE RETRIEVAL (please share it)
I am checking different opensource vector databases to create a database of images. This database will be quite dynamic at the beginning, and it should be open for changes (CRUD) and reindexing for the full time of the project, which can be some years.
I see that Milvus and Weaviate are two potentially good candidates, but I would love to know more from people who had already experience with them.
Does somebody have experience with those vector DBs or other potential candidates for a image retrieval application?
Eine neue Folge #arthistoCast ist online!
@KlusikEckert spricht mit @peterbell und Stefanie Schneider über das visuelle Suchen in großen Bilddatenmengen. Dabei geht es neben einer Reflexion über unsere Suchstrategien in der Kunstgeschichte auch um Prototypen für das visuelle Suchen.
Unter
https://www.arthistoricum.net/themen/podcasts/arthistocast
Und überall, wo es Podcasts gibt!
#digitaleKunstgeschichte #wissenschaftspodcast #wisskomm #computervision #openai #imgsai #iArt #imageretrieval #informationretrieval
IB Teguh TM
Harald Klinke
KjR
#arthistoCast