PyTorchKR Papers

최신 논문

총 2,066편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2007.13172공식 코드 · Papers with Code

Learning and aggregating deep local descriptors for instance-level recognition

We propose an efficient method to learn deep local descriptors for instance-level recognition. The training only requires examples of positive and negative image pairs and is performed as metric learning of sum-pooled global image descriptors. At inference, the local descriptors are provided by the activations of internal components of the network. We demonstrate why such an approach learns local descriptors that work well for image similarity estimation with classical efficient match kernel methods. The experimental validation studies the trade-off between performance and memory requirements of the state-of-the-art image search approach based on match kernels. Compared to existing local descriptors, the proposed ones perform better in two instance-level recognition tasks and keep memory requirements lower. We experimentally show that global descriptors are not effective enough at large scale and that local descriptors are essential. We achieve state-of-the-art performance, in some cases even with a backbone network as small as ResNet18.

arXiv:2007.12212공식 코드 · Papers with Code

ZSCRGAN: A GAN-based Expectation Maximization Model for Zero-Shot Retrieval of Images from Textual Descriptions

Most existing algorithms for cross-modal Information Retrieval are based on a supervised train-test setup, where a model learns to align the mode of the query (e.g., text) to the mode of the documents (e.g., images) from a given training set. Such a setup assumes that the training set contains an exhaustive representation of all possible classes of queries. In reality, a retrieval model may need to be deployed on previously unseen classes, which implies a zero-shot IR setup. In this paper, we propose a novel GAN-based model for zero-shot text to image retrieval. When given a textual description as the query, our model can retrieve relevant images in a zero-shot setup. The proposed model is trained using an Expectation-Maximization framework. Experiments on multiple benchmark datasets show that our proposed model comfortably outperforms several state-of-the-art zero-shot text to image retrieval models, as well as zero-shot classification and hashing models suitably used for retrieval.

arXiv:2007.12163

Smooth-AP: Smoothing the Path Towards Large-Scale Image Retrieval

Optimising a ranking-based metric, such as Average Precision (AP), is notoriously challenging due to the fact that it is non-differentiable, and hence cannot be optimised directly using gradient-descent methods. To this end, we introduce an objective that optimises instead a smoothed approximation of AP, coined Smooth-AP. Smooth-AP is a plug-and-play objective function that allows for end-to-end training of deep networks with a simple and elegant implementation. We also present an analysis for why directly optimising the ranking based metric of AP offers benefits over other deep metric learning losses. We apply Smooth-AP to standard retrieval benchmarks: Stanford Online products and VehicleID, and also evaluate on larger-scale datasets: INaturalist for fine-grained category retrieval, and VGGFace2 and IJB-C for face retrieval. In all cases, we improve the performance over the state-of-the-art, especially for larger-scale datasets, thus demonstrating the effectiveness and scalability of Smooth-AP to real-world scenarios.

arXiv:2007.10915

Wireless Image Retrieval at the Edge

We study the image retrieval problem at the wireless edge, where an edge device captures an image, which is then used to retrieve similar images from an edge server. These can be images of the same person or a vehicle taken from other cameras at different times and locations. Our goal is to maximize the accuracy of the retrieval task under power and bandwidth constraints over the wireless link. Due to the stringent delay constraint of the underlying application, sending the whole image at a sufficient quality is not possible. We propose two alternative schemes based on digital and analog communications, respectively. In the digital approach, we first propose a deep neural network (DNN) aided retrieval-oriented image compression scheme, whose output bit sequence is transmitted over the channel using conventional channel codes. In the analog joint source and channel coding (JSCC) approach, the feature vectors are directly mapped into channel symbols. We evaluate both schemes on image based re-identification (re-ID) tasks under different channel conditions, including both static and fading channels. We show that the JSCC scheme significantly increases the end-to-end accuracy, speeds up the encoding process, and provides graceful degradation with channel conditions. The proposed architecture is evaluated through extensive simulations on different datasets and channel conditions, as well as through ablation studies.

arXiv:2007.09464

A Bag of Visual Words Model for Medical Image Retrieval

Medical Image Retrieval is a challenging field in Visual information retrieval, due to the multi-dimensional and multi-modal context of the underlying content. Traditional models often fail to take the intrinsic characteristics of data into consideration, and have thus achieved limited accuracy when applied to medical images. The Bag of Visual Words (BoVW) is a technique that can be used to effectively represent intrinsic image features in vector space, so that applications like image classification and similar-image search can be optimized. In this paper, we present a MedIR approach based on the BoVW model for content-based medical image retrieval. As medical images as multi-dimensional, they exhibit underlying cluster and manifold information which enhances semantic relevance and allows for label uniformity. Hence, the BoVW features extracted for each image are used to train a supervised machine learning classifier based on positive and negative training images, for extending content based image retrieval. During experimental validation, the proposed model performed very well, achieving a Mean Average Precision of 88.89% during top-3 image retrieval experiments.

arXiv:2007.09433

Volumetric Transformer Networks

Existing techniques to encode spatial invariance within deep convolutional neural networks (CNNs) apply the same warping field to all the feature channels. This does not account for the fact that the individual feature channels can represent different semantic parts, which can undergo different spatial transformations w.r.t. a canonical configuration. To overcome this limitation, we introduce a learnable module, the volumetric transformer network (VTN), that predicts channel-wise warping fields so as to reconfigure intermediate CNN features spatially and channel-wisely. We design our VTN as an encoder-decoder network, with modules dedicated to letting the information flow across the feature channels, to account for the dependencies between the semantic parts. We further propose a loss function defined between the warped features of pairs of instances, which improves the localization ability of VTN. Our experiments show that VTN consistently boosts the features' representation power and consequently the networks' accuracy on fine-grained image recognition and instance-level image retrieval.

arXiv:2007.09162

Improving Object Detection with Selective Self-supervised Self-training

We study how to leverage Web images to augment human-curated object detection datasets. Our approach is two-pronged. On the one hand, we retrieve Web images by image-to-image search, which incurs less domain shift from the curated data than other search methods. The Web images are diverse, supplying a wide variety of object poses, appearances, their interactions with the context, etc. On the other hand, we propose a novel learning method motivated by two parallel lines of work that explore unlabeled data for image classification: self-training and self-supervised learning. They fail to improve object detectors in their vanilla forms due to the domain gap between the Web images and curated datasets. To tackle this challenge, we propose a selective net to rectify the supervision signals in Web images. It not only identifies positive bounding boxes but also creates a safe zone for mining hard negative boxes. We report state-of-the-art results on detecting backpacks and chairs from everyday scenes, along with other challenging object classes.

arXiv:2007.09699공식 코드 · Papers with Code

Improving the HardNet Descriptor

In the thesis we consider the problem of local feature descriptor learning for wide baseline stereo focusing on the HardNet descriptor, which is close to state-of-the-art. AMOS Patches dataset is introduced, which improves robustness to illumination and appearance changes. It is based on registered images from selected cameras from the AMOS dataset. We provide recommendations on the patch dataset creation process and evaluate HardNet trained on data of different modalities. We also introduce a dataset combination and reduction methods, that allow comparable performance on a significantly smaller dataset. HardNet8, consistently outperforming the original HardNet, benefits from the architectural choices made: connectivity pattern, final pooling, receptive field, CNN building blocks found by manual or automatic search algorithms -- DARTS. We show impact of overlooked hyperparameters such as batch size and length of training on the descriptor quality. PCA dimensionality reduction further boosts performance and also reduces memory footprint. Finally, the insights gained lead to two HardNet8 descriptors: one performing well on a variety of benchmarks -- HPatches, AMOS Patches and IMW Phototourism, the other is optimized for IMW Phototourism.

arXiv:2007.08628

Universal Model for Multi-Domain Medical Image Retrieval

Medical Image Retrieval (MIR) helps doctors quickly find similar patients' data, which can considerably aid the diagnosis process. MIR is becoming increasingly helpful due to the wide use of digital imaging modalities and the growth of the medical image repositories. However, the popularity of various digital imaging modalities in hospitals also poses several challenges to MIR. Usually, one image retrieval model is only trained to handle images from one modality or one source. When there are needs to retrieve medical images from several sources or domains, multiple retrieval models need to be maintained, which is cost ineffective. In this paper, we study an important but unexplored task: how to train one MIR model that is applicable to medical images from multiple domains? Simply fusing the training data from multiple domains cannot solve this problem because some domains become over-fit sooner when trained together using existing methods. Therefore, we propose to distill the knowledge in multiple specialist MIR models into a single multi-domain MIR model via universal embedding to solve this problem. Using skin disease, x-ray, and retina image datasets, we validate that our proposed universal model can effectively accomplish multi-domain MIR.

arXiv:2007.08019

Attention-Based Query Expansion Learning

Query expansion is a technique widely used in image search consisting in combining highly ranked images from an original query into an expanded query that is then reissued, generally leading to increased recall and precision. An important aspect of query expansion is choosing an appropriate way to combine the images into a new query. Interestingly, despite the undeniable empirical success of query expansion, ad-hoc methods with different caveats have dominated the landscape, and not a lot of research has been done on learning how to do query expansion. In this paper we propose a more principled framework to query expansion, where one trains, in a discriminative manner, a model that learns how images should be aggregated to form the expanded query. Within this framework, we propose a model that leverages a self-attention mechanism to effectively learn how to transfer information between the different images before aggregating them. Our approach obtains higher accuracy than existing approaches on standard benchmarks. More importantly, our approach is the only one that consistently shows high accuracy under different regimes, overcoming caveats of existing methods.

arXiv:2007.07081

Nodule2vec: a 3D Deep Learning System for Pulmonary Nodule Retrieval Using Semantic Representation

Content-based retrieval supports a radiologist decision making process by presenting the doctor the most similar cases from the database containing both historical diagnosis and further disease development history. We present a deep learning system that transforms a 3D image of a pulmonary nodule from a CT scan into a low-dimensional embedding vector. We demonstrate that such a vector representation preserves semantic information about the nodule and offers a viable approach for content-based image retrieval (CBIR). We discuss the theoretical limitations of the available datasets and overcome them by applying transfer learning of the state-of-the-art lung nodule detection model. We evaluate the system using the LIDC-IDRI dataset of thoracic CT scans. We devise a similarity score and show that it can be utilized to measure similarity 1) between annotations of the same nodule by different radiologists and 2) between the query nodule and the top four CBIR results. A comparison between doctors and algorithm scores suggests that the benefit provided by the system to the radiologist end-user is comparable to obtaining a second radiologist's opinion.

arXiv:2007.06624

A new approach to descriptors generation for image retrieval by analyzing activations of deep neural network layers

In this paper, we consider the problem of descriptors construction for the task of content-based image retrieval using deep neural networks. The idea of neural codes, based on fully connected layers activations, is extended by incorporating the information contained in convolutional layers. It is known that the total number of neurons in the convolutional part of the network is large and the majority of them have little influence on the final classification decision. Therefore, in the paper we propose a novel algorithm that allows us to extract the most significant neuron activations and utilize this information to construct effective descriptors. The descriptors consisting of values taken from both the fully connected and convolutional layers perfectly represent the whole image content. The images retrieved using these descriptors match semantically very well to the query image, and also they are similar in other secondary image characteristics, like background, textures or color distribution. These features of the proposed descriptors are verified experimentally based on the IMAGENET1M dataset using the VGG16 neural network.

arXiv:2007.07177

MosAIc: Finding Artistic Connections across Culture with Conditional Image Retrieval

We introduce MosAIc, an interactive web app that allows users to find pairs of semantically related artworks that span different cultures, media, and millennia. To create this application, we introduce Conditional Image Retrieval (CIR) which combines visual similarity search with user supplied filters or "conditions". This technique allows one to find pairs of similar images that span distinct subsets of the image corpus. We provide a generic way to adapt existing image retrieval data-structures to this new domain and provide theoretical bounds on our approach's efficiency. To quantify the performance of CIR systems, we introduce new datasets for evaluating CIR methods and show that CIR performs non-parametric style transfer. Finally, we demonstrate that our CIR data-structures can identify "blind spots" in Generative Adversarial Networks (GAN) where they fail to properly model the true data distribution.

arXiv:2007.05299

Data-Efficient Ranking Distillation for Image Retrieval

Recent advances in deep learning has lead to rapid developments in the field of image retrieval. However, the best performing architectures incur significant computational cost. Recent approaches tackle this issue using knowledge distillation to transfer knowledge from a deeper and heavier architecture to a much smaller network. In this paper we address knowledge distillation for metric learning problems. Unlike previous approaches, our proposed method jointly addresses the following constraints i) limited queries to teacher model, ii) black box teacher model with access to the final output representation, and iii) small fraction of original training data without any ground-truth labels. In addition, the distillation method does not require the student and teacher to have same dimensionality. Addressing these constraints reduces computation requirements, dependency on large-scale training datasets and addresses practical scenarios of limited or partial access to private data such as teacher models or the corresponding training data/labels. The key idea is to augment the original training set with additional samples by performing linear interpolation in the final output representation space. Distillation is then performed in the joint space of original and augmented teacher-student sample representations. Results demonstrate that our approach can match baseline models trained with full supervision. In low training sample settings, our approach outperforms the fully supervised approach on two challenging image retrieval datasets, ROxford5k and RParis6k \cite{Roxf} with the least possible teacher supervision.

arXiv:1208.6335

Comparative Study and Optimization of Feature-Extraction Techniques for Content based Image Retrieval

The aim of a Content-Based Image Retrieval (CBIR) system, also known as Query by Image Content (QBIC), is to help users to retrieve relevant images based on their contents. CBIR technologies provide a method to find images in large databases by using unique descriptors from a trained image. The image descriptors include texture, color, intensity and shape of the object inside an image. Several feature-extraction techniques viz., Average RGB, Color Moments, Co-occurrence, Local Color Histogram, Global Color Histogram and Geometric Moment have been critically compared in this paper. However, individually these techniques result in poor performance. So, combinations of these techniques have also been evaluated and results for the most efficient combination of techniques have been presented and optimized for each class of image query. We also propose an improvement in image retrieval performance by introducing the idea of Query modification through image cropping. It enables the user to identify a region of interest and modify the initial query to refine and personalize the image retrieval results.

arXiv:2007.03375

Location Sensitive Image Retrieval and Tagging

People from different parts of the globe describe objects and concepts in distinct manners. Visual appearance can thus vary across different geographic locations, which makes location a relevant contextual information when analysing visual data. In this work, we address the task of image retrieval related to a given tag conditioned on a certain location on Earth. We present LocSens, a model that learns to rank triplets of images, tags and coordinates by plausibility, and two training strategies to balance the location influence in the final ranking. LocSens learns to fuse textual and location information of multimodal queries to retrieve related images at different levels of location granularity, and successfully utilizes location information to improve image tagging.

arXiv:2007.00145

Modality-Agnostic Attention Fusion for visual search with text feedback

Image retrieval with natural language feedback offers the promise of catalog search based on fine-grained visual features that go beyond objects and binary attributes, facilitating real-world applications such as e-commerce. Our Modality-Agnostic Attention Fusion (MAAF) model combines image and text features and outperforms existing approaches on two visual search with modifying phrase datasets, Fashion IQ and CSS, and performs competitively on a dataset with only single-word modifications, Fashion200k. We also introduce two new challenging benchmarks adapted from Birds-to-Words and Spot-the-Diff, which provide new settings with rich language inputs, and we show that our approach without modification outperforms strong baselines. To better understand our model, we conduct detailed ablations on Fashion IQ and provide visualizations of the surprising phenomenon of words avoiding "attending" to the image region they refer to.

arXiv:2006.16331공식 코드 · Papers with Code

Asymmetric metric learning for knowledge transfer

Knowledge transfer from large teacher models to smaller student models has recently been studied for metric learning, focusing on fine-grained classification. In this work, focusing on instance-level image retrieval, we study an asymmetric testing task, where the database is represented by the teacher and queries by the student. Inspired by this task, we introduce asymmetric metric learning, a novel paradigm of using asymmetric representations at training. This acts as a simple combination of knowledge transfer with the original metric learning task. We systematically evaluate different teacher and student models, metric learning and knowledge transfer loss functions on the new asymmetric testing as well as the standard symmetric testing task, where database and queries are represented by the same model. We find that plain regression is surprisingly effective compared to more complex knowledge transfer mechanisms, working best in asymmetric testing. Interestingly, our asymmetric metric learning approach works best in symmetric testing, allowing the student to even outperform the teacher.

arXiv:2006.13330

A Mean-Field Theory for Learning the Schönberg Measure of Radial Basis Functions

We develop and analyze a projected particle Langevin optimization method to learn the distribution in the Sch\"{o}nberg integral representation of the radial basis functions from training samples. More specifically, we characterize a distributionally robust optimization method with respect to the Wasserstein distance to optimize the distribution in the Sch\"{o}nberg integral representation. To provide theoretical performance guarantees, we analyze the scaling limits of a projected particle online (stochastic) optimization method in the mean-field regime. In particular, we prove that in the scaling limits, the empirical measure of the Langevin particles converges to the law of a reflected It\^{o} diffusion-drift process. Moreover, the drift is also a function of the law of the underlying process. Using It\^{o} lemma for semi-martingales and Grisanov's change of measure for the Wiener processes, we then derive a Mckean-Vlasov type partial differential equation (PDE) with Robin boundary conditions that describes the evolution of the empirical measure of the projected Langevin particles in the mean-field regime. In addition, we establish the existence and uniqueness of the steady-state solutions of the derived PDE in the weak sense. We apply our learning approach to train radial kernels in the kernel locally sensitive hash (LSH) functions, where the training data-set is generated via a $k$-mean clustering method on a small subset of data-base. We subsequently apply our kernel LSH with a trained kernel for image retrieval task on MNIST data-set, and demonstrate the efficacy of our kernel learning approach. We also apply our kernel learning approach in conjunction with the kernel support vector machines (SVMs) for classification of benchmark data-sets.

arXiv:2006.13046

Rotation Invariant Deep CBIR

Introduction of Convolutional Neural Networks has improved results on almost every image-based problem and Content-Based Image Retrieval is not an exception. But the CNN features, being rotation invariant, creates problems to build a rotation-invariant CBIR system. Though rotation-invariant features can be hand-engineered, the retrieval accuracy is very low because by hand engineering only low-level features can be created, unlike deep learning models that create high-level features along with low-level features. This paper shows a novel method to build a rotational invariant CBIR system by introducing a deep learning orientation angle detection model along with the CBIR feature extraction model. This paper also highlights that this rotation invariant deep CBIR can retrieve images from a large dataset in real-time.