• Corpus ID: 6628301

Fracking Deep Convolutional Image Descriptors

@article{SimoSerra2014FrackingDC,
  title={Fracking Deep Convolutional Image Descriptors},
  author={Edgar Simo-Serra and Eduard Trulls and Luis Ferraz and Iasonas Kokkinos and Francesc Moreno-Noguer},
  journal={ArXiv},
  year={2014},
  volume={abs/1412.6537}
}
In this paper we propose a novel framework for learning local image descriptors in a discriminative manner. [] Key Method We propose to explore this space with a stochastic sampling of the training set, in combination with an aggressive mining strategy over both the positive and negative samples which we denote as "fracking". We perform a thorough evaluation of the architecture hyper-parameters, and demonstrate large performance gains compared to both standard CNN learning strategies, hand-crafted image…
NetVLAD: CNN Architecture for Weakly Supervised Place Recognition
TLDR
A convolutional neural network architecture that is trainable in an end-to-end manner directly for the place recognition task, and significantly outperforms non-learnt image representations and off-the-shelf CNN descriptors on two challenging place recognition benchmarks.
Object Retrieval with Deep Convolutional Features
TLDR
This chapter recommends a simple pipeline for encoding the local activations of a convolutional layer of a pretrained CNN utilizing the well-known Bag of Words (BoW) aggregation scheme and called bag of local Convolutional features (BLCF).
Fine-Tuning CNN Image Retrieval with No Human Annotation
TLDR
It is shown that both hard-positive and hard-negative examples, selected by exploiting the geometry and the camera positions available from the 3D models, enhance the performance of particular-object retrieval.
NetVLAD: CNN Architecture for Weakly Supervised Place Recognition
TLDR
A convolutional neural network architecture that is trainable in an end-to-end manner directly for the place recognition task and an efficient training procedure which can be applied on very large-scale weakly labelled tasks are developed.
Trainable Siamese keypoint descriptors for real-time applications
TLDR
This paper explores a Siamese pairing of fully connected neural networks for the purpose of learning discriminative local feature descriptors, and demonstrates consistent speedup as compared to such state-of-the-art methods as SIFT and FREAK on PCs as well as in embedded systems.
Geometric Deep Learning for Shape Analysis
TLDR
This thesis pioneers the generalization of CNNs to discrete manifolds and provides an alternative formulation of the spectral convolution operation in terms of the windowed Fourier transform to overcome the drawbacks of the Fourier one.
Object instance identification with fully convolutional networks
TLDR
A novel approach for instance search and object detection, applied to museum visits, that relies on fully convolutional networks (FCN) to obtain region proposals and object representation and it is shown that the descriptors obtained using the proposed network outperform those from previous state-of-the-art approaches.
Video Retrieval of Specific Persons in Specific Locations
This thesis explores good practices for improving the detection of specific people in specific places, as defined in the TRECVID 2016 benchmark for instance search in large scale video datasets. An
A Survey on Deep Visual Place Recognition
TLDR
A survey of the state-of-the-art of research on visual place recognition, focusing on how it has been shaped by the recent advances in deep learning, is presented.
Local Convolutional Features with Unsupervised Training for Image Retrieval
TLDR
A deep convolutional architecture that yields patch-level descriptors, called Patch-CKN, as an alternative to the popular SIFT descriptor for image retrieval, which yields competitive results compared to supervised CNN alternatives on patch and image retrieval.
...
...

References

SHOWING 1-10 OF 36 REFERENCES
Learned local descriptors for recognition and matching
TLDR
This work investigates the idea of considering keypoint description as a discriminative classification task, and proposes to learn descriptors as it enables, in contrast to hand-crafted descriptors, to adjust the amount of invariance and distinctiveness to the task at hand.
Learning Local Feature Descriptors Using Convex Optimisation
TLDR
It is shown that learning the pooling regions for the descriptor can be formulated as a convex optimisation problem selecting the regions using sparsity, and an extension of the learning formulations to a weakly supervised case, which allows us to learn the descriptors from unannotated image collections.
Discriminative Learning of Local Image Descriptors
TLDR
A set of building blocks for constructing descriptors which can be combined together and jointly optimized so as to minimize the error of a nearest-neighbor classifier are described.
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
TLDR
This paper proposes a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%.
LDAHash: Improved Matching with Smaller Descriptors
TLDR
This work reduces the size of the descriptors by representing them as short binary strings and learn descriptor invariance from examples, and shows extensive experimental validation, demonstrating the advantage of the proposed approach.
ImageNet classification with deep convolutional neural networks
TLDR
A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.
Boosting Binary Keypoint Descriptors
TLDR
A novel framework to learn an extremely compact binary descriptor called Bin Boost that is very robust to illumination and viewpoint changes and significantly outperforms the state-of-the-art binary descriptors and performs similarly to the best floating-point descriptors at a fraction of the matching time and memory footprint.
Convolutional Neural Networks Learn Compact Local Image Descriptors
TLDR
It is shown emprically that a newly suggested loss formulation allows a Convolutional Neural Network to find compact local image descriptors that perform comparably to state-of-the-art approaches.
Iterative Quantization: A Procrustean Approach to Learning Binary Codes for Large-Scale Image Retrieval
TLDR
This paper addresses the problem of learning similarity-preserving binary codes for efficient similarity search in large-scale image collections by proposing a simple and efficient alternating minimization algorithm, dubbed iterative quantization (ITQ), and demonstrating an application of ITQ to learning binary attributes or "classemes" on the ImageNet data set.
A performance evaluation of local descriptors
TLDR
It is observed that the ranking of the descriptors is mostly independent of the interest region detector and that the SIFT-based descriptors perform best and Moments and steerable filters show the best performance among the low dimensional descriptors.
...
...