We propose an attentive local feature descriptor suitable for large-scale image retrieval, referred to as DELF (DEep Local Feature). The new feature is based on convolutional neural networks, which are trained only with image-level annotations on a landmark image dataset. To identify semantically useful local features for image retrieval, we also propose an attention mechanism for keypoint selection, which shares most network layers with the descriptor. This framework can be used for image retrieval as a drop-in replacement for other keypoint detectors and descriptors, enabling more accurate feature matching and geometric verification. Our system produces reliable confidence scores to reject false positives---in particular, it is robust against queries that have no correct match in the database. To evaluate the proposed descriptor, we introduce a new large-scale dataset, referred to as Google-Landmarks dataset, which involves challenges in both database and query such as background clutter, partial occlusion, multiple landmarks, objects in variable scales, etc. We show that DELF outperforms the state-of-the-art global and local descriptors in the large-scale setting by significant margins. Code and dataset can be found at the project webpage: https://github.com/tensorflow/models/tree/master/research/delf .
Retrieving object instances among cluttered scenes efficiently requires compact yet comprehensive regional image representations. Intuitively, object semantics can help build the index that focuses on the most relevant regions. However, due to the lack of bounding-box datasets for objects of interest among retrieval benchmarks, most recent work on regional representations has focused on either uniform or class-agnostic region selection. In this paper, we first fill the void by providing a new dataset of landmark bounding boxes, based on the Google Landmarks dataset, that includes $86k$ images with manually curated boxes from $15k$ unique landmarks. Then, we demonstrate how a trained landmark detector, using our new dataset, can be leveraged to index image regions and improve retrieval accuracy while being much more efficient than existing regional methods. In addition, we introduce a novel regional aggregated selective match kernel (R-ASMK) to effectively combine information from detected regions into an improved holistic image representation. R-ASMK boosts image retrieval accuracy substantially with no dimensionality increase, while even outperforming systems that index image regions independently. Our complete image retrieval system improves upon the previous state-of-the-art by significant margins on the Revisited Oxford and Paris datasets. Code and data available at the project webpage: https://github.com/tensorflow/models/tree/master/research/delf.
Recent studies in image retrieval task have shown that ensembling different models and combining multiple global descriptors lead to performance improvement. However, training different models for ensemble is not only difficult but also inefficient with respect to time or memory. In this paper, we propose a novel framework that exploits multiple global descriptors to get an ensemble-like effect while it can be trained in an end-to-end manner. The proposed framework is flexible and expandable by the global descriptor, CNN backbone, loss, and dataset. Moreover, we investigate the effectiveness of combining multiple global descriptors with quantitative and qualitative analysis. Our extensive experiments show that the combined descriptor outperforms a single global descriptor, as it can utilize different types of feature properties. In the benchmark evaluation, the proposed framework achieves the state-of-the-art performance on the CARS196, CUB200-2011, In-shop Clothes and Stanford Online Products on image retrieval tasks by a large margin compared to competing approaches. Our model implementations and pretrained models are publicly available.
Person re-identification (PReID) has received increasing attention due to it is an important part in intelligent surveillance. Recently, many state-of-the-art methods on PReID are part-based deep models. Most of them focus on learning the part feature representation of person body in horizontal direction. However, the feature representation of body in vertical direction is usually ignored. Besides, the spatial information between these part features and the different feature channels is not considered. In this study, we introduce a multi-branches deep model for PReID. Specifically, the model consists of five branches. Among the five branches, two of them learn the local feature with spatial information from horizontal or vertical orientations, respectively. The other one aims to learn interdependencies knowledge between different feature channels generated by the last convolution layer. The remains of two other branches are identification and triplet sub-networks, in which the discriminative global feature and a corresponding measurement can be learned simultaneously. All the five branches can improve the representation learning. We conduct extensive comparative experiments on three PReID benchmarks including CUHK03, Market-1501 and DukeMTMC-reID. The proposed deep framework outperforms many state-of-the-art in most cases.
Recently it has shown that the policy-gradient methods for reinforcement learning have been utilized to train deep end-to-end systems on natural language processing tasks. What's more, with the complexity of understanding image content and diverse ways of describing image content in natural language, image captioning has been a challenging problem to deal with. To the best of our knowledge, most state-of-the-art methods follow a pattern of sequential model, such as recurrent neural networks (RNN). However, in this paper, we propose a novel architecture for image captioning with deep reinforcement learning to optimize image captioning tasks. We utilize two networks called "policy network" and "value network" to collaboratively generate the captions of images. The experiments are conducted on Microsoft COCO dataset, and the experimental results have verified the effectiveness of the proposed method.
Image-level feature descriptors obtained from convolutional neural networks have shown powerful representation capabilities for image retrieval. In this paper, we present an unsupervised method to aggregate deep convolutional features into compact yet discriminative image vectors by simulating the dynamics of heat diffusion. A distinctive problem in image retrieval is that repetitive or bursty features tend to dominate feature representations, leading to less than ideal matches. We show that by considering each deep feature as a heat source, our method is able to avoiding over-representation of bursty features. We additionally provide a practical solution for the proposed aggregation method, which is further demonstrated in our experimental evaluation. Finally, we extensively evaluate the proposed approach with pre-trained and fine-tuned deep networks on common public benchmarks, and show superior performance compared to previous work.
Image-level feature descriptors obtained from convolutional neural networks have shown powerful representation capabilities for image retrieval. In this paper, we present an unsupervised method to aggregate deep convolutional features into compact yet discriminative image vectors by simulating the dynamics of heat diffusion. A distinctive problem in image retrieval is that repetitive or bursty features tend to dominate feature representations, leading to less than ideal matches. We show that by leveraging elegant properties of the heat equation, our method is able to select informative features while avoiding over-representation of bursty features. We additionally present a theoretical time complexity analysis showing the efficiency of our method, which is further demonstrated in our experimental evaluation. Finally, we extensively evaluate the proposed approach with pre-trained and fine-tuned deep networks on common public benchmarks, and show superior performance compared to previous work.
Person re-identification is being widely used in the forensic, and security and surveillance system, but person re-identification is a challenging task in real life scenario. Hence, in this work, a new feature descriptor model has been proposed using a multilayer framework of Gaussian distribution model on pixel features, which include color moments, color space values and Schmid filter responses. An image of a person usually consists of distinct body regions, usually with differentiable clothing followed by local colors and texture patterns. Thus, the image is evaluated locally by dividing the image into overlapping regions. Each region is further fragmented into a set of local Gaussians on small patches. A global Gaussian encodes, these local Gaussians for each region creating a multi-level structure. Hence, the global picture of a person is described by local level information present in it, which is often ignored. Also, we have analyzed the efficiency of earlier metric learning methods on this descriptor. The performance of the descriptor is evaluated on four public available challenging datasets and the highest accuracy achieved on these datasets are compared with similar state-of-the-arts, which demonstrate the superior performance.
We investigate the problem of automatically determining what type of shoe left an impression found at a crime scene. This recognition problem is made difficult by the variability in types of crime scene evidence (ranging from traces of dust or oil on hard surfaces to impressions made in soil) and the lack of comprehensive databases of shoe outsole tread patterns. We find that mid-level features extracted by pre-trained convolutional neural nets are surprisingly effective descriptors for this specialized domains. However, the choice of similarity measure for matching exemplars to a query image is essential to good performance. For matching multi-channel deep features, we propose the use of multi-channel normalized cross-correlation and analyze its effectiveness. Our proposed metric significantly improves performance in matching crime scene shoeprints to laboratory test impressions. We also show its effectiveness in other cross-domain image retrieval problems: matching facade images to segmentation labels and aerial photos to map images. Finally, we introduce a discriminatively trained variant and fine-tune our system through our proposed metric, obtaining state-of-the-art performance.
An interactive image retrieval system learns which images in the database belong to a user's query concept, by analyzing the example images and feedback provided by the user. The challenge is to retrieve the relevant images with minimal user interaction. In this work, we propose to solve this problem by posing it as a binary classification task of classifying all images in the database as being relevant or irrelevant to the user's query concept. Our method combines active learning with graph-based semi-supervised learning (GSSL) to tackle this problem. Active learning reduces the number of user interactions by querying the labels of the most informative points and GSSL allows to use abundant unlabeled data along with the limited labeled data provided by the user. To efficiently find the most informative point, we use an uncertainty sampling based method that queries the label of the point nearest to the decision boundary of the classifier. We estimate this decision boundary using our heuristic of adaptive threshold. To utilize huge volumes of unlabeled data we use an efficient approximation based method that reduces the complexity of GSSL from $O(n^3)$ to $O(n)$, making GSSL scalable. We make the classifier robust to the diversity and noisy labels associated with images in large databases by incorporating information from multiple modalities such as visual information extracted from deep learning based models and semantic information extracted from the WordNet. High F1 scores within few relevance feedback rounds in our experiments with concepts defined on AnimalWithAttributes and Imagenet (1.2 million images) datasets indicate the effectiveness and scalability of our approach.
Employing part-level features for pedestrian image description offers fine-grained information and has been verified as beneficial for person retrieval in very recent literature. A prerequisite of part discovery is that each part should be well located. Instead of using external cues, e.g., pose estimation, to directly locate parts, this paper lays emphasis on the content consistency within each part. Specifically, we target at learning discriminative part-informed features for person retrieval and make two contributions. (i) A network named Part-based Convolutional Baseline (PCB). Given an image input, it outputs a convolutional descriptor consisting of several part-level features. With a uniform partition strategy, PCB achieves competitive results with the state-of-the-art methods, proving itself as a strong convolutional baseline for person retrieval. (ii) A refined part pooling (RPP) method. Uniform partition inevitably incurs outliers in each part, which are in fact more similar to other parts. RPP re-assigns these outliers to the parts they are closest to, resulting in refined parts with enhanced within-part consistency. Experiment confirms that RPP allows PCB to gain another round of performance boost. For instance, on the Market-1501 dataset, we achieve (77.4+4.2)% mAP and (92.3+1.5)% rank-1 accuracy, surpassing the state of the art by a large margin.