Filters








441 Hits in 1.3 sec

Robust Feature Bundling [chapter]

Stefan Romberg, Moritz August, Christian X. Ries, Rainer Lienhart
2012 Lecture Notes in Computer Science  
In this work we present a feature bundling technique that aggregates individual local features with features from their spatial neighborhood into bundles. The resulting bundles carry more information of the underlying image content than single visual words. As in practice an exact search for such bundles is infeasible, we employ a robust approximate similarity search with min-hashing in order to retrieve images containing similar bundles. We demonstrate the benefits of these bundles for small
more » ... ject retrieval, i.e. logo recognition, and generic image retrieval. Multiple bundling strategies are explored and thoroughly evaluated on three different datasets.
doi:10.1007/978-3-642-34778-8_5 fatcat:om2c6oqllfbwxpberdodsswibm

Bundle min-Hashing

Stefan Romberg, Rainer Lienhart
2013 International Journal of Multimedia Information Retrieval  
recognition resultsBold values denote the best score test set including logo and logo-free images (3,960 images) is then used to compute the classification scores.Results Method Precision Recall Romberg  ... 
doi:10.1007/s13735-013-0040-x fatcat:3jzzxkfvnjbizieidjusag2kyu

Multimodal Image Retrieval

Stefan Romberg, Rainer Lienhart, Eva Hörster
2012 International Journal of Multimedia Information Retrieval  
In this work, we extend the standard single-layer probabilistic Latent Semantic Analysis (pLSA) (Hofmann in Mach Learn 42(1-2):177-196, 2001) to multiple layers. As multiple layers should naturally handle multiple modalities and a hierarchy of abstractions, we denote this new approach multilayer multimodal probabilistic Latent Semantic Analysis (mm-pLSA). We derive the training and inference rules for the smallest possible non-degenerated mm-pLSA model: a model with two leaf-pLSAs and a single
more » ... op-level pLSA node merging the two leaf-pLSAs. We evaluate this approach on two pairs of different modalities: SIFT features and image annotations (tags) as well as the combination of SIFT and HOG features. We also propose a fast and strictly stepwise forward procedure to initialize the bottom-up mm-pLSA model, which in turn can then be post-optimized by the general mm-pLSA learning algorithm. The proposed approach is evaluated in a query-by-example retrieval task where various variants of our mm-pLSA system are compared to systems relying on a single modality and other ad-hoc combinations of feature histograms. We further describe possible pitfalls of the mm-pLSA training and analyze the resulting model yielding an intuitive explanation of its behaviour.
doi:10.1007/s13735-012-0006-4 fatcat:usa7xvlgrzfdxoenf3ovs6hqle

Towards universal visual vocabularies

Christian X. Ries, Stefan Romberg, Rainer Lienhart
2010 2010 IEEE International Conference on Multimedia and Expo  
Many content-based image mining systems extract local features from images to obtain an image description based on discrete feature occurrences. Such applications require a visual vocabulary also known as visual codebook or visual dictionary to discretize the extracted high-dimensional features to visual words in an efficient yet accurate way. Once such an application operates on images of a very specific domain the question arises if a vocabulary built from those domain-specific images needs
more » ... be used or if a "universal" visual vocabulary can be used instead. A universal visual vocabulary may be computed from images of a different domain once and then be re-used for various applications and other domains. We therefore evaluate several visual vocabularies from different image domains by determining their performance at pLSA-based image classification on several datasets. We empirically conclude that vocabularies suit our classification tasks equally well disregarding the image domain they were derived from.
doi:10.1109/icme.2010.5583878 dblp:conf/icmcs/RiesRL10 fatcat:ttjr3rxpandcnjn5v4wf5gfi6m

Bundle min-hashing for logo recognition

Stefan Romberg, Rainer Lienhart
2013 Proceedings of the 3rd ACM conference on International conference on multimedia retrieval - ICMR '13  
Precision Recall Romberg et al. [16] 0.98 0.61 Revaud et al. [14] ≥ Testing.  ... 
doi:10.1145/2461466.2461486 dblp:conf/mir/RombergL13 fatcat:tcio7g3mwfedrkq26l4fwgywmu

Multimodal pLSA on visual features and tags

Stefan Romberg, Eva Horster, Rainer Lienhart
2009 2009 IEEE International Conference on Multimedia and Expo  
This work studies a new approach for image retrieval on largescale community databases. Our proposed system explores two different modalities: visual features and communitygenerated metadata, such as tags. We use topic models to derive a high-level representation appropriate for retrieval for each of our images in the database. We evaluate the proposed approach experimentally in a query-by-example retrieval task and compare our results to systems relying solely on visual features or tag
more » ... . It is shown that the proposed multimodal system outperforms the unimodal systems by approximately 36%.
doi:10.1109/icme.2009.5202522 dblp:conf/icmcs/RombergHL09 fatcat:kfremgkw3bgq3kn3mriwutqa3q

Multilayer pLSA for multimodal image retrieval

Rainer Lienhart, Stefan Romberg, Eva Hörster
2009 Proceeding of the ACM International Conference on Image and Video Retrieval - CIVR '09  
It is current state of knowledge that our neocortex consists of six layers [10] . We take this knowledge from neuroscience as an inspiration to extend the standard single-layer probabilistic Latent Semantic Analysis (pLSA) [13] to multiple layers. As multiple layers should naturally handle multiple modalities and a hierarchy of abstractions, we denote this new approach multilayer multimodal probabilistic Latent Semantic Analysis (mm-pLSA). We derive the training and inference rules for the
more » ... est possible non-degenerated mm-pLSA model: a model with two leaf-pLSAs (here from two different data modalities: image tags and visual image features) and a single top-level pLSA node merging the two leaf-pLSAs. From this derivation it is obvious how to extend the learning and inference rules to more modalities and more layers. We also propose a fast and strictly stepwise forward procedure to initialize bottom-up the mm-pLSA model, which in turn can then be post-optimized by the general mm-pLSA learning algorithm. We evaluate the proposed approach experimentally in a query-by-example retrieval task using 50dimensional topic vectors as image models. We compare various variants of our mm-pLSA system to systems relying solely on visual features or tag features and analyze possible pitfalls of the mm-pLSA training. It is shown that the best variant of the the proposed mm-pLSA system outperforms the unimodal systems by approximately 19% in our query-by-example task.
doi:10.1145/1646396.1646408 dblp:conf/civr/LienhartRH09 fatcat:akhc5cbl3faflj54btxlfor4py

Scalable logo recognition in real-world images

Stefan Romberg, Lluis Garcia Pueyo, Rainer Lienhart, Roelof van Zwol
2011 Proceedings of the 1st ACM International Conference on Multimedia Retrieval - ICMR '11  
In this paper we propose a highly effective and scalable framework for recognizing logos in images. At the core of our approach lays a method for encoding and indexing the relative spatial layout of local features detected in the logo images. Based on the analysis of the local features and the composition of basic spatial structures, such as edges and triangles, we can derive a quantized representation of the regions in the logos and minimize the false positive detections. Furthermore, we
more » ... e a cascaded index for scalable multi-class recognition of logos. For the evaluation of our system, we have constructed and released a logo recognition benchmark which consists of manually labeled logo images, complemented with nonlogo images, all posted on Flickr. The dataset consists of a training, validation, and test set with 32 logo-classes. We thoroughly evaluate our system with this benchmark and show that our approach effectively recognizes different logo classes with high precision.
doi:10.1145/1991996.1992021 dblp:conf/mir/RombergPLZ11 fatcat:hbxbayl4tbaztnqcermylamhpq

Multimodal ranking for image search on community databases

Fabian Richter, Stefan Romberg, Eva Hörster, Rainer Lienhart
2010 Proceedings of the international conference on Multimedia information retrieval - MIR '10  
Searching for relevant images given a query term is an important task in nowadays large-scale community databases. The image ranking approach presented in this work represents an image collection as a graph that is built using a multimodal similarity measure based on visual features and user tags. We perform a random walk on this graph to find the most common images. Further we discuss several scalability issues of the proposed approach and show how in this framework queries can be answered
more » ... . Experimental results validate the effectiveness of the presented algorithm.
doi:10.1145/1743384.1743402 dblp:conf/mir/RichterRHL10 fatcat:ylplqntbq5go7nm25wrcfvi35i

Improving VLAD: Hierarchical coding and a refined local coordinate system

Christian Eggert, Stefan Romberg, Rainer Lienhart
2014 2014 IEEE International Conference on Image Processing (ICIP)  
The enormous growth of image databases calls for new techniques for fast and effective image search that scales with millions of images. Most importantly, the setting requires a compact but also descriptive image signature. Recently, the vector of aggregated local descriptors (VLAD) [1] has received much attention in large-scale image retrieval. In this paper we present two modifications for VLAD which improve the retrieval performance of the signature.
doi:10.1109/icip.2014.7025610 dblp:conf/icip/EggertRL14 fatcat:4ydpmxkbibfqrczvh7hbuo4mxy

Partial contour matching for document pieces with content-based prior

Fabian Richter, Christian X. Ries, Stefan Romberg, Rainer Lienhart
2014 2014 IEEE International Conference on Multimedia and Expo (ICME)  
In this paper we present a method for aligning shredded document pieces based on outer contours and content-based prior information. Our approach relies on domain-specific knowledge that document pieces must complement each other when aligned correctly. Building on this intuition we propose a variant of MSAC (M-estimator SAmple Consensus) to estimate an hypothesis that recovers the spatial relationship between pairs of pieces. To do so we first approximate their boundaries by polygons from
more » ... we define consensus sets between fragments. Each consensus set provides multiple hypotheses for aligning one piece onto the other. An optimal hypothesis is identified by applying a two-stage procedure in which we discard locally inconsistent hypotheses before verifying the remainder for global consistency.
doi:10.1109/icme.2014.6890237 dblp:conf/icmcs/RichterRRL14 fatcat:pwccmmghxnegfhajhz2d4g2zgy

Potency testing of veterinary vaccines: The way from in vivo to in vitro

Judith Romberg, Stefan Lang, Elisabeth Balks, Elisabeth Kamphuis, Karin Duchow, Daniela Loos, Henriette Rau, Andreas Motitschke, Carmen Jungbäck
2012 Biologicals (Print)  
For further information please go to: www.pei.de www.iabs.org www.edqm.eu Before being placed on the market inactivated vaccines are predominantly tested in vivo, mainly in laboratory animals. In recent years, substantial efforts have been made either to modify these animal tests in order to reduce the number of required animals and the stress imposed on them (Refinement) or to completely replace these experiments by in-vitro tests. The acceptance of these tests differs considerably between
more » ... ine manufacturers and licensing authorities. It is thus comprehensible that vaccine manufacturers hesitate to adopt the new test methods. The aim of the meeting is to enhance the acceptance of the new test methods by all licensing authorities (Europe/USA) and their application by all vaccine manufacturers. Scientific committee Carmen Jungbäck PEI, Germany
doi:10.1016/j.biologicals.2011.10.004 pmid:22075457 fatcat:es7l6cngivcbtahq4jdpfv7vki

HHU at SemEval-2019 Task 6: Context Does Matter - Tackling Offensive Language Identification and Categorization with ELMo

Alexander Oberstrass, Julia Romberg, Anke Stoll, Stefan Conrad
2019 Proceedings of the 13th International Workshop on Semantic Evaluation  
We present our results for OffensEval: Identifying and Categorizing Offensive Language in Social Media (SemEval 2019 -Task 6). Our results show that context embeddings are important features for the three different subtasks in connection with classical machine and with deep learning. Our best model reached place 3 of 75 in sub-task B with a macro F 1 of 0.719. Our approaches for sub-task A and C perform less well but could also deliver promising results.
doi:10.18653/v1/s19-2112 dblp:conf/semeval/OberstrassRS019 fatcat:jxlpst5wsfb4fj25kiuqnf7za4

HHU at SemEval-2017 Task 5: Fine-Grained Sentiment Analysis on Financial Data using Machine Learning Methods

Tobias Cabanski, Julia Romberg, Stefan Conrad
2017 Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017)  
In this Paper a system for solving SemEval-2017 Task 5 is presented. This task is divided into two tracks where the sentiment of microblog messages and news headlines has to be predicted. Since two submissions were allowed, two different machine learning methods were developed to solve this task, a support vector machine approach and a recurrent neural network approach. To feed in data for these approaches, different feature extraction methods are used, mainly word representations and lexica.
more » ... e best submissions for both tracks are provided by the recurrent neural network which achieves a score of 0.729 in track 1 and 0.702 in track 2.
doi:10.18653/v1/s17-2141 dblp:conf/semeval/CabanskiRC17 fatcat:oeoi7gp5nvdgfpwnj4pli4xbse

Digital Contact Tracing Service: An improved decentralized design for privacy and effectiveness [article]

Kilian Holzapfel, Martina Karl, Linus Lotz, Georg Carle, Christian Djeffal, Christian Fruck, Christian Haack, Dirk Heckmann, Philipp H. Kindt, Michael Köppl, Patrick Krause, Lolian Shtembari (+25 others)
2020 arXiv   pre-print
We propose a decentralized digital contact tracing service that preserves the users' privacy by design while complying to the highest security standards. Our approach is based on Bluetooth and measures actual encounters of people, the contact time period, and estimates the proximity of the contact. We trace the users' contacts and the possible spread of infectious diseases while preventing location tracking of users, protecting their data and identity. We verify and improve the impact of
more » ... g based on epidemiological models. We compare a centralized and decentralized approach on a legal perspective and find a decentralized approach preferable considering proportionality and data minimization.
arXiv:2006.16960v1 fatcat:mse4iov7bja5tkxjlxsgdgl5le
« Previous Showing results 1 — 15 out of 441 results