@InProceedings{8553380,
  author = {A. Jose and T. Horstmann and J. Ohm},
  booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},
  title = {Optimized Binary Hashing Codes Generated by Siamese Neural Networks for Image Retrieval},
  year = {2018},
  pages = {1487-1491},
  abstract = {In this paper, we use a Siamese Neural Network based hashing method for generating binary codes with certain properties. The training architecture takes a pair of images as input. The loss function trains the network so that similar images are mapped to similar binary codes and dissimilar images to different binary codes. We add additional constraints in form of loss functions that enforce certain properties on the binary codes. The main motivation of incorporating the first constraint is maximization of entropy by generating binary codes with the same number of 1s and Os. The second constraint minimizes the mutual information between binary codes by generating orthogonal binary codes for dissimilar images. For this, we introduce orthogonality criterion for binary codes consisting of the binary values 0 and 1. Furthermore, we evaluate the properties such as mutual information and entropy of the binary codes generated with the additional constraints. We also analyze the influence of different bit sizes on those properties. The retrieval performance is evaluated by measuring Mean Average Precision (MAP) values and the results are compared with other state-of-the-art approaches.},
  keywords = {binary codes;entropy;file organisation;image coding;image retrieval;learning (artificial intelligence);neural nets;optimisation;Siamese Neural networks;similar binary codes;dissimilar images;orthogonal binary codes;optimized binary hashing codes;entropy maximization;mutual information;image retrieval;Binary codes;Training;Entropy;Mutual information;Neural networks;Image retrieval;Europe;Siamese Neural Networks;Binary Hashing;Image Retrieval;Code Property Training;Information Theoretic Criteria},
  doi = {10.23919/EUSIPCO.2018.8553380},
  issn = {2076-1465},
  month = {Sep.},
  url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439081.pdf},
}

{"_id":"XDcueRfJTp8wFm5ZQ","bibbaseid":"jose-horstmann-ohm-optimizedbinaryhashingcodesgeneratedbysiameseneuralnetworksforimageretrieval-2018","authorIDs":[],"author_short":["Jose, A.","Horstmann, T.","Ohm, J."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["A."],"propositions":[],"lastnames":["Jose"],"suffixes":[]},{"firstnames":["T."],"propositions":[],"lastnames":["Horstmann"],"suffixes":[]},{"firstnames":["J."],"propositions":[],"lastnames":["Ohm"],"suffixes":[]}],"booktitle":"2018 26th European Signal Processing Conference (EUSIPCO)","title":"Optimized Binary Hashing Codes Generated by Siamese Neural Networks for Image Retrieval","year":"2018","pages":"1487-1491","abstract":"In this paper, we use a Siamese Neural Network based hashing method for generating binary codes with certain properties. The training architecture takes a pair of images as input. The loss function trains the network so that similar images are mapped to similar binary codes and dissimilar images to different binary codes. We add additional constraints in form of loss functions that enforce certain properties on the binary codes. The main motivation of incorporating the first constraint is maximization of entropy by generating binary codes with the same number of 1s and Os. The second constraint minimizes the mutual information between binary codes by generating orthogonal binary codes for dissimilar images. For this, we introduce orthogonality criterion for binary codes consisting of the binary values 0 and 1. Furthermore, we evaluate the properties such as mutual information and entropy of the binary codes generated with the additional constraints. We also analyze the influence of different bit sizes on those properties. The retrieval performance is evaluated by measuring Mean Average Precision (MAP) values and the results are compared with other state-of-the-art approaches.","keywords":"binary codes;entropy;file organisation;image coding;image retrieval;learning (artificial intelligence);neural nets;optimisation;Siamese Neural networks;similar binary codes;dissimilar images;orthogonal binary codes;optimized binary hashing codes;entropy maximization;mutual information;image retrieval;Binary codes;Training;Entropy;Mutual information;Neural networks;Image retrieval;Europe;Siamese Neural Networks;Binary Hashing;Image Retrieval;Code Property Training;Information Theoretic Criteria","doi":"10.23919/EUSIPCO.2018.8553380","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439081.pdf","bibtex":"@InProceedings{8553380,\n author = {A. Jose and T. Horstmann and J. Ohm},\n booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},\n title = {Optimized Binary Hashing Codes Generated by Siamese Neural Networks for Image Retrieval},\n year = {2018},\n pages = {1487-1491},\n abstract = {In this paper, we use a Siamese Neural Network based hashing method for generating binary codes with certain properties. The training architecture takes a pair of images as input. The loss function trains the network so that similar images are mapped to similar binary codes and dissimilar images to different binary codes. We add additional constraints in form of loss functions that enforce certain properties on the binary codes. The main motivation of incorporating the first constraint is maximization of entropy by generating binary codes with the same number of 1s and Os. The second constraint minimizes the mutual information between binary codes by generating orthogonal binary codes for dissimilar images. For this, we introduce orthogonality criterion for binary codes consisting of the binary values 0 and 1. Furthermore, we evaluate the properties such as mutual information and entropy of the binary codes generated with the additional constraints. We also analyze the influence of different bit sizes on those properties. The retrieval performance is evaluated by measuring Mean Average Precision (MAP) values and the results are compared with other state-of-the-art approaches.},\n keywords = {binary codes;entropy;file organisation;image coding;image retrieval;learning (artificial intelligence);neural nets;optimisation;Siamese Neural networks;similar binary codes;dissimilar images;orthogonal binary codes;optimized binary hashing codes;entropy maximization;mutual information;image retrieval;Binary codes;Training;Entropy;Mutual information;Neural networks;Image retrieval;Europe;Siamese Neural Networks;Binary Hashing;Image Retrieval;Code Property Training;Information Theoretic Criteria},\n doi = {10.23919/EUSIPCO.2018.8553380},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439081.pdf},\n}\n\n","author_short":["Jose, A.","Horstmann, T.","Ohm, J."],"key":"8553380","id":"8553380","bibbaseid":"jose-horstmann-ohm-optimizedbinaryhashingcodesgeneratedbysiameseneuralnetworksforimageretrieval-2018","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570439081.pdf"},"keyword":["binary codes;entropy;file organisation;image coding;image retrieval;learning (artificial intelligence);neural nets;optimisation;Siamese Neural networks;similar binary codes;dissimilar images;orthogonal binary codes;optimized binary hashing codes;entropy maximization;mutual information;image retrieval;Binary codes;Training;Entropy;Mutual information;Neural networks;Image retrieval;Europe;Siamese Neural Networks;Binary Hashing;Image Retrieval;Code Property Training;Information Theoretic Criteria"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2018url.bib","creationDate":"2021-02-13T15:38:40.408Z","downloads":0,"keywords":["binary codes;entropy;file organisation;image coding;image retrieval;learning (artificial intelligence);neural nets;optimisation;siamese neural networks;similar binary codes;dissimilar images;orthogonal binary codes;optimized binary hashing codes;entropy maximization;mutual information;image retrieval;binary codes;training;entropy;mutual information;neural networks;image retrieval;europe;siamese neural networks;binary hashing;image retrieval;code property training;information theoretic criteria"],"search_terms":["optimized","binary","hashing","codes","generated","siamese","neural","networks","image","retrieval","jose","horstmann","ohm"],"title":"Optimized Binary Hashing Codes Generated by Siamese Neural Networks for Image Retrieval","year":2018,"dataSources":["yiZioZximP7hphDpY","iuBeKSmaES2fHcEE9"]}