Re-Ranking for Writer Identification and Writer Retrieval

Jordan, Simon; Seuret, Mathias; Král, Pavel; Lenc, Ladislav; Martínek, Jiří; Wiermann, Barbara; Schwinger, Tobias; Maier, Andreas; Christlein, Vincent

doi:10.1007/978-3-030-57058-3_40

Cited by 7 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the best of our knowledge, Jordan et al. in [9] are the only ones who employed this scenario. However, their approach differs from ours since they used query expansion to add more positive examples before evaluation using ESVMS.…”

Section: Discussionmentioning

confidence: 99%

“…In [9] Jordan et al experimented with reciprocal relationships in two ways. First, they integrated them into the Jaccard distance and computed the final rankings based on a weighted sum of the original distance and the Jaccard distance.…”

Section: Related Workmentioning

confidence: 99%

“…While it is quite common in image retrieval methods to use re-ranking strategies, it has not gained much attention in the field of writer identification and writer retrieval. To the best of our knowledge, Jordan et al in [9] are the only ones who employed this scenario. However, their approach differs from ours since they used query expansion to add more positive examples before evaluation using ESVMS.…”

Section: Influence Of Different Parts Of the Pipelinementioning

confidence: 99%

See 2 more Smart Citations

Writer identification and writer retrieval based on NetVLAD with Re‐ranking

Rasoulzadeh

BabaAli

2021

IET Biometrics

View full text Add to dashboard Cite

The issue of writer identification and writer retrieval, which is considered a challenging problem in the field of document analysis and recognition is addressed here.. A novel pipeline is proposed for the problem at hand by employing a unified neural network architecture consisting of the ResNet-20 as a feature extractor and an integrated NetV-LAD layer, inspired by the vector of locally aggregated descriptors (VLAD), in the head of the latter part. Having defined this architecture, the triplet semi-hard loss function is used to directly learn an embedding for individual input image patches. Subsequently, the generalised max-pooling technique is employed for the aggregation of embedded descriptors of each handwritten image. Also, a novel re-ranking strategy is introduced for the task of identification and retrieval based on the k-reciprocal nearest neighbours, and it is shown that the pipeline can benefit tremendously from this step. Experimental evaluation has been done on the three publicly available datasets: the ICDAR 2013, CVL, and KHATT datasets. Results indicate that while the performance was comparable to the state-of-the-art KHATT, the writer identification and writer retrieval pipeline achieve superior performance on the ICDAR 2013 and CVL datasets in terms of mAP.This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Influence Of Different Parts Of the Pipelinementioning

confidence: 99%

See 1 more Smart Citation

Writer identification and writer retrieval based on NetVLAD with Re‐ranking

Rasoulzadeh

BabaAli

2021

IET Biometrics

View full text Add to dashboard Cite

show abstract

“…Consider that our proposed pipeline is mostly inspired by the works of Christlein et al [4,7,9] and Jordan et. al [21]. However, with our proposed pipeline consisting of the unified neural network architecture with the NetVLAD layer, and re-ranking strategy based query expansion and k-reciprocal nearest neighbors, we could improve upon the state-of-the-art on the ICDAR13 and CVL datasets.…”

Section: Related Workmentioning

confidence: 99%

“…A common approach in order to boost performance of information retrieval systems is automatic query expansion (QE) [21]. With an initial ranked lists L(q) computed, query expansion reformulates each query sample q and obtains the improved ranked list by requering using the newly formed query instead of q.…”

Section: Query Expansion and K-reciporal Nearest Neighborsmentioning

confidence: 99%

Writer Identification and Writer Retrieval Based on NetVLAD with Re-ranking

Rasoulzadeh,

Babaali

2020

Preprint

View full text Add to dashboard Cite

This paper addresses writer identification and retrieval which is a challenging problem in the document analysis field. In this work, a novel pipeline is proposed for the problem by employing a unified neural network architecture consisting of the ResNet-20 as a feature extractor and an integrated NetVLAD layer, inspired by the vectors of locally aggregated descriptors (VLAD), in the head of the latter part. Having defined this architecture, triplet semi-hard loss function is used to directly learn an embedding for individual input image patches. Generalised max-pooling is used for the aggregation of embedded descriptors of each handwritten image. In the evaluation part, for identification and retrieval, re-ranking has been done based on query expansion and k-reciprocal nearest neighbours, and it is shown that the pipeline can benefit tremendously from this step. Experimental evaluation shows that our writer identification and writer retrieval pipeline is superior compared to the state-of-the-art pipelines, as our results on the publicly available ICDAR13 and CVL datasets set new standards by achieving 96.5% and 98.4% mAP, respectively.

show abstract

Self-supervised Vision Transformers with Data Augmentation Strategies Using Morphological Operations for Writer Retrieval

Peer

Kleber

Sablatnig

2022

Frontiers in Handwriting Recognition

View full text Add to dashboard Cite

This paper presents an unsupervised approach for writer retrieval based on clustering SIFT descriptors detected at keypoint locations resulting in pseudo-cluster labels. With those cluster labels, a residual network followed by our proposed NetRVLAD, an encoding layer with reduced complexity compared to NetVLAD, is trained on 32×32 patches at keypoint locations. Additionally, we suggest a graph-based reranking algorithm called SGR to exploit similarities of the page embeddings to boost the retrieval performance. Our approach is evaluated on two historical datasets (Historical-WI and HisIR19). We include an evaluation of different backbones and NetRVLAD. It competes with related work on historical datasets without using explicit encodings. We set a new State-of-the-art on both datasets by applying our reranking scheme and show that our approach achieves comparable performance on a modern dataset as well.

show abstract

Re-Ranking for Writer Identification and Writer Retrieval

Cited by 7 publications

References 24 publications

Writer identification and writer retrieval based on NetVLAD with Re‐ranking

Writer identification and writer retrieval based on NetVLAD with Re‐ranking

Writer Identification and Writer Retrieval Based on NetVLAD with Re-ranking

Self-supervised Vision Transformers with Data Augmentation Strategies Using Morphological Operations for Writer Retrieval

Contact Info

Product

Resources

About