Novel Similarity Metric Learning Using Deep Learning and Root SIFT for Person Re-identification

Vidhyalakshmi, M. K.; Poovammal, E.; Bhaskar, Vidhyacharan; Sathyanarayanan, J.

doi:10.1007/s11277-020-07948-1

Cited by 7 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Besides, the fusion method is varied from a simple concatenation to more sophisticated attention mechanisms [ 18 , 23 , 24 ]. Additionally, the previous dual-stream logic is modified by redoubling each stream and implementing a Siamese scheme [ 25 ]. Additionally, the hybrid CNN and SIFT methods were evaluated using sequence-modelling tasks to capture video dynamics in opposition to an optical flow [ 26 , 27 ].…”

Section: Related Workmentioning

confidence: 99%

SIFT-CNN: When Convolutional Neural Networks Meet Dense SIFT Descriptors for Image and Sequence Classification

et al. 2022

View full text Add to dashboard Cite

Despite the success of hand-crafted features in computer visioning for many years, nowadays, this has been replaced by end-to-end learnable features that are extracted from deep convolutional neural networks (CNNs). Whilst CNNs can learn robust features directly from image pixels, they require large amounts of samples and extreme augmentations. On the contrary, hand-crafted features, like SIFT, exhibit several interesting properties as they can provide local rotation invariance. In this work, a novel scheme combining the strengths of SIFT descriptors with CNNs, namely SIFT-CNN, is presented. Given a single-channel image, one SIFT descriptor is computed for every pixel, and thus, every pixel is represented as an M-dimensional histogram, which ultimately results in an M-channel image. Thus, the SIFT image is generated from the SIFT descriptors for all the pixels in a single-channel image, while at the same time, the original spatial size is preserved. Next, a CNN is trained to utilize these M-channel images as inputs by operating directly on the multiscale SIFT images with the regular convolution processes. Since these images incorporate spatial relations between the histograms of the SIFT descriptors, the CNN is guided to learn features from local gradient information of images that otherwise can be neglected. In this manner, the SIFT-CNN implicitly acquires a local rotation invariance property, which is desired for problems where local areas within the image can be rotated without affecting the overall classification result of the respective image. Some of these problems refer to indirect immunofluorescence (IIF) cell image classification, ground-based all-sky image-cloud classification and human lip-reading classification. The results for the popular datasets related to the three different aforementioned problems indicate that the proposed SIFT-CNN can improve the performance and surpasses the corresponding CNNs trained directly on pixel values in various challenging tasks due to its robustness in local rotations. Our findings highlight the importance of the input image representation in the overall efficiency of a data-driven system.

show abstract

Section: Related Workmentioning

confidence: 99%

SIFT-CNN: When Convolutional Neural Networks Meet Dense SIFT Descriptors for Image and Sequence Classification

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Dengan menerapkan Novel Similartiy Matric Learning yang menggabungkan metode deep features dan Root Scale Invariant Features Transform (Root SIFT) dapat meningkatkan keberhasil dalam pencocokan gambar. Setelah dilakukan eskperimen oleh Vidhyalakshmi dkk, diperoleh persentase keberhasilan pencocokan mencapai 74,45% pada dataset CUHK 03 [10].…”

Section: Metode Similarity Metric Learningunclassified

Design of the Attendance System using RFID and Similarity Metric Learning at Universitas Dinamika Bangsa

Toscany

Rahim

Bustami

et al. 2022

ijcs

View full text Add to dashboard Cite

Admission of new students at Universitas Dinamika Bangsa has a significant increase, so that the number of active students on campus becomes more. This has an impact on the recapitulation process of lecture absence by study staff to be longer than before, due to the calculation process that is done manually. This research will create a system designed to increase efficiency in lecture attendance activities by presenting an attendance system equipped with an RFID (Radio Frequency Identification) Reader and camera. In addition to the process of digitization of attendance data will be added similiarity metric learning method for validation of student absences that are done periodically. The method to be used in system development is prototyping. This research resulted in a system design with UML modeling and absentee device design.

show abstract

“…In 2019, Rodríguez et al [ 21 ] proposed a CNN-driven patch descriptor that captures affine invariance that is based on the first stages of SIFT. In 2021, Vidhyalakshmi et al [ 22 ] combined the Root SIFT descriptor and CNN feature to enhance the matching performance for the person re-identification problem.…”

Section: Introductionmentioning

confidence: 99%

A Cognitive Sample Consensus Method for the Stitching of Drone-Based Aerial Images Supported by a Generative Adversarial Network for False Positive Reduction

Seo

2022

Sensors

View full text Add to dashboard Cite

When using drone-based aerial images for panoramic image generation, the unstableness of the shooting angle often deteriorates the quality of the resulting image. To prevent these polluting effects from affecting the stitching process, this study proposes deep learning-based outlier rejection schemes that apply the architecture of the generative adversarial network (GAN) to reduce the falsely estimated hypothesis relating to a transform produced by a given baseline method, such as the random sample consensus method (RANSAC). To organize the training dataset, we obtain rigid transforms to resample the images via the operation of RANSAC for the correspondences produced by the scale-invariant feature transform descriptors. In the proposed method, the discriminator of GAN makes a pre-judgment of whether the estimated target hypothesis sample produced by RANSAC is true or false, and it recalls the generator to confirm the authenticity of the discriminator’s inference by comparing the differences between the generated samples and the target sample. We have tested the proposed method for drone-based aerial images and some miscellaneous images. The proposed method has been shown to have relatively stable and good performances even in receiver-operated tough conditions.

show abstract

Novel Similarity Metric Learning Using Deep Learning and Root SIFT for Person Re-identification

Cited by 7 publications

References 15 publications

SIFT-CNN: When Convolutional Neural Networks Meet Dense SIFT Descriptors for Image and Sequence Classification

SIFT-CNN: When Convolutional Neural Networks Meet Dense SIFT Descriptors for Image and Sequence Classification

Design of the Attendance System using RFID and Similarity Metric Learning at Universitas Dinamika Bangsa

A Cognitive Sample Consensus Method for the Stitching of Drone-Based Aerial Images Supported by a Generative Adversarial Network for False Positive Reduction

Contact Info

Product

Resources

About