Scalable Discrete Supervised Multimedia Hash Learning With Clustering

Zhang, Shifeng; Li, Jianmin; Jiang, Mengqing; Yuan, Pingzhi; Zhang, Bo

doi:10.1109/tcsvt.2017.2710345

Cited by 23 publications

(12 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…User Relevance Feedback has been used since the 1960s to improve queries for information retrieval [24] and saw a boom in the 1990s and 2000s for multimedia retrieval [11,12,29,[33][34][35]. Later, it started fading as hash based approaches [20], product quantization [13], and deep learning models [7,32,40] were proven more efficient for retrieval on large-scale collections. However, in recent years the issue of scalability has largely been resolved and the state-of-the-art URF systems for large scale multimedia retrieval are competitive with other approaches and require fewer examples to train their models than the supervised approaches [16,18,36].…”

Section: Related Workmentioning

confidence: 99%

Impact of Interaction Strategies on User Relevance Feedback

Khan

Jónsson

Zahálka

et al. 2021

Proceedings of the 2021 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

User Relevance Feedback (URF) is a class of interactive learning methods that rely on the interaction between a human user and a system to analyze a media collection. To improve URF system evaluation and design better systems, it is important to understand the impact that different interaction strategies can have. Based on the literature and observations from real user sessions from the Lifelog Search Challenge and Video Browser Showdown, we analyze interaction strategies related to (a) labeling positive and negative examples, and (b) applying filters based on users' domain knowledge. Experiments show that there is no single optimal labeling strategy, as the best strategy depends on both the collection and the task. In particular, our results refute the common assumption that providing more training examples is always beneficial: strategies with a smaller number of prototypical examples lead to better results in some cases. We further observe that while expert filtering is unsurprisingly beneficial, aggressive filtering, especially by novice users, can hinder the completion of tasks. Finally, we observe that combining URF with filters leads to better results than using filters alone. CCS CONCEPTS• Information systems → Multimedia and multimodal retrieval; Evaluation of retrieval results.

show abstract

Section: Related Workmentioning

confidence: 99%

Impact of Interaction Strategies on User Relevance Feedback

Khan

Jónsson

Zahálka

et al. 2021

Proceedings of the 2021 International Conference on Multimedia Retrieval

View full text Add to dashboard Cite

show abstract

“…We compare our SCDH method with recent state-of-the-art deep hashing methods, including pairwise based methods such as DSH [31], DHN [64], DPSH [28], DQN [4], DISH [63], DSDH [27], triplet based methods like NINH [22], FTDE [65], BOH [6], DRLIH [61], and unary loss based methods like CNNBH [13], SSDH [59]. They follow similar experimental settings, but different methods may use different deep networks, thus we train on several types of network (AlexNet, VGGNet, ResNet, etc.)…”

Section: B Comparison On Supervised Hashingmentioning

confidence: 99%

Semantic Cluster Unary Loss for Efficient Deep Hashing

Zhang

2019

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

Hashing method maps similar data to binary hashcodes with smaller hamming distance, which has received broad attention due to its low storage cost and fast retrieval speed. With the rapid development of deep learning, deep hashing methods have achieved promising results in efficient information retrieval. Most existing deep hashing methods adopt pairwise or triplet losses to deal with similarities underlying the data, but their training are difficult and less efficient because O(n 2 ) data pairs and O(n 3 ) triplets are involved. To address these issues, we propose a novel deep hashing algorithm with unary loss which can be trained very efficiently. First of all, we introduce a Unary Upper Bound of the traditional triplet loss, thus reducing the complexity to O(n) and bridging the classificationbased unary loss and the triplet loss. Second, we propose a novel Semantic Cluster Deep Hashing (SCDH) algorithm by introducing a modified Unary Upper Bound loss, named Semantic Cluster Unary Loss (SCUL). The resultant hashcodes form several compact clusters, which means hashcodes in the same cluster have similar semantic information. We also demonstrate that the proposed SCDH is easy to be extended to semi-supervised settings by incorporating the state-of-the-art semi-supervised learning algorithms. Experiments on large-scale datasets show that the proposed method is superior to state-of-the-art hashing algorithms.

show abstract

“…Hashing methods can map similar data to similar binary codes that have small Hamming distance, due to the low storage cost and fast retrieval speed, hashing methods have been receiving broad attention [1]. Hashing methods have also been used in many applications, such as image retrieval [2]- [10] and video retrieval [11]- [13]. Traditional hashing methods [6], [7], [14], [15] take pre-extracted image features as input, and then learn hash functions by exploiting the data structures or applying the similarity preserving regularizations.…”

Section: Introductionmentioning

confidence: 99%

SSDH: Semi-Supervised Deep Hashing for Large Scale Image Retrieval

Zhang¹,

Peng²

2019

IEEE Trans. Circuits Syst. Video Technol.

119

View full text Add to dashboard Cite

Abstract-Hashing methods have been widely used for efficient similarity retrieval on large scale image database. Traditional hashing methods learn hash functions to generate binary codes from hand-crafted features, which achieve limited accuracy since the hand-crafted features cannot optimally represent the image content and preserve the semantic similarity. Recently, several deep hashing methods have shown better performance because the deep architectures generate more discriminative feature representations. However, these deep hashing methods are mainly designed for supervised scenarios, which only exploit the semantic similarity information, but ignore the underlying data structures. In this paper, we propose the semi-supervised deep hashing (SSDH) approach, to perform more effective hash function learning by simultaneously preserving semantic similarity and underlying data structures. The main contributions are as follows: (1) We propose a semi-supervised loss to jointly minimize the empirical error on labeled data, as well as the embedding error on both labeled and unlabeled data, which can preserve the semantic similarity and capture the meaningful neighbors on the underlying data structures for effective hashing. (2) A semisupervised deep hashing network is designed to extensively exploit both labeled and unlabeled data, in which we propose an online graph construction method to benefit from the evolving deep features during training to better capture semantic neighbors. To the best of our knowledge, the proposed deep network is the first deep hashing method that can perform hash code learning and feature learning simultaneously in a semi-supervised fashion. Experimental results on 5 widely-used datasets show that our proposed approach outperforms the state-of-the-art hashing methods.

show abstract

Scalable Discrete Supervised Multimedia Hash Learning With Clustering

Cited by 23 publications

References 35 publications

Impact of Interaction Strategies on User Relevance Feedback

Impact of Interaction Strategies on User Relevance Feedback

Semantic Cluster Unary Loss for Efficient Deep Hashing

SSDH: Semi-Supervised Deep Hashing for Large Scale Image Retrieval

Contact Info

Product

Resources

About