Nearest Neighbor Retrieval Using Distance-Based Hashing

Athitsos, Vassilis; Potamias, Michalis; Papapetrou, Panagiotis; Kollios, George

doi:10.1109/icde.2008.4497441

Cited by 73 publications

(62 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We obtain significantly higher pruning than that provided by distance based hashing [3] methods, trained on our distance function.…”

Section: Main Contributions Of the Papermentioning

confidence: 95%

“…The challenge in developing such an access method is two-fold. Firstly, indexing using such distances has not been well-studied till date -the recently proposed distance based hashing [3] performs dataset pruning for arbitrary distances. Secondly, video fingerprints are generally of high dimension and varying length.…”

Section: Fi J=1mentioning

confidence: 99%

“…Since our proposed distance function is non-metric, LSH cannot be used in our setup as the locality sensitive property holds only for metric distances. Instead, we have experimented with the recently proposed distance based hashing (DBH) [3] scheme, which can be used for arbitrary distance measures.…”

Section: Global Image Featuresmentioning

confidence: 99%

See 2 more Smart Citations

Efficient and Robust Detection of Duplicate Videos in a Large Database

Sarkar

Singh

Ghosh

et al. 2010

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

We present an efficient and accurate method for duplicate video detection in a large database using video fingerprints. We have empirically chosen the Color Layout Descriptor, a compact and robust frame-based descriptor, to create fingerprints which are further encoded by vector quantization. We propose a new non-metric distance measure to find the similarity between the query and a database video fingerprint and experimentally show its superior performance over other distance measures for accurate duplicate detection. Efficient search can not be performed for high dimensional data using a non-metric distance measure with existing indexing techniques. Therefore, we develop novel search algorithms based on pre-computed distances and new dataset pruning techniques yielding practical retrieval times. We perform experiments with a database of 38000 videos, worth 1600 hours of content.For individual queries with an average duration of 60 sec (about 50% of the average database video length), the duplicate video is retrieved in 0.032 sec, on Intel Xeon with CPU 2.33GHz, with a very high accuracy of 97.5%.

show abstract

“…We obtain significantly higher pruning than that provided by distance based hashing [3] methods, trained on our distance function.…”

Section: Main Contributions Of the Papermentioning

confidence: 95%

Section: Fi J=1mentioning

confidence: 99%

See 1 more Smart Citation

Efficient and Robust Detection of Duplicate Videos in a Large Database

Sarkar

Singh

Ghosh

et al. 2010

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…However, the asymmetric relationship between kNN and RkNN makes it hard to adapt techniques for kNN to RkNN. A simple adaptation of a 1NN scheme described in [4] would necessitate more hash functions and require scanning each bucket to which at least one kNN candidate of each RkNN candidate maps to. This would still lead to an approximate answer,the kNN method being approximate.…”

Section: Related Workmentioning

confidence: 99%

“…Another related work is the distance based hashing [4] used for 1NN computation under non-metric similarity measures. However, the asymmetric relationship between kNN and RkNN makes it hard to adapt techniques for kNN to RkNN.…”

Section: Related Workmentioning

confidence: 99%

Efficient RkNN retrieval with arbitrary non-metric similarity measures

Deepak

Deshpande

2010

Proc. VLDB Endow.

View full text Add to dashboard Cite

A RkNN query returns all objects whose nearest k neighbors contain the query object. In this paper, we consider RkNN query processing in the case where the distances between attribute values are not necessarily metric. Dissimilarities between objects could then be a monotonic aggregate of dissimilarities between their values, such aggregation functions being specified at query time. We outline real world cases that motivate RkNN processing in such scenarios. We consider the AL-Tree index and its applicability in RkNN query processing. We develop an approach that exploits the group level reasoning enabled by the AL-Tree in RkNN processing. We evaluate our approach against a Naive approach that performs sequential scans on contiguous data and an improved block-based approach that we provide. We use real-world datasets and synthetic data with varying characteristics for our experiments. This extensive empirical evaluation shows that our approach is better than existing methods in terms of computational and disk access costs, leading to significantly better response times.

show abstract