Camera Style Research Articles

A GAN-Based Self-Training Framework for Unsupervised Domain Adaptive Person Re-Identification.

As a crucial task in surveillance and security, person re-identification (re-ID) aims to identify the targeted pedestrians across multiple images captured by non-overlapping cameras. However, existing person re-ID solutions have two main challenges: the lack of pedestrian identification labels in the captured images, and domain shift issue between different domains. A generative adversarial networks (GAN)-based self-training framework with progressive augmentation (SPA) is proposed to obtain the robust features of the unlabeled data from the target domain, according to the preknowledge of the labeled data from the source domain. Specifically, the proposed framework consists of two stages: the style transfer stage (STrans), and self-training stage (STrain). First, the targeted data is complemented by a camera style transfer algorithm in the STrans stage, in which CycleGAN and Siamese Network are integrated to preserve the unsupervised self-similarity (the similarity of the same image between before and after transformation) and domain dissimilarity (the dissimilarity between a transferred source image and the targeted image). Second, clustering and classification are alternately applied to enhance the model performance progressively in the STrain stage, in which both global and local features of the target-domain images are obtained. Compared with the state-of-the-art methods, the proposed method achieves the competitive accuracy on two existing datasets.

Cross-view similarity exploration for unsupervised cross-domain person re-identification

Due to the existence of a domain gap between different domains, when a model trained on one domain is applied to other domain, performance will drop dramatically. For the moment, some of the solutions are concentrating on reducing data distribution discrepancy in different domains, but they ignore unlabeled samples in the target domain. To address this problem, we propose the cross-view similarity exploration (CVSE) method, which combines style-transferred samples to optimize the CNN model and the relationship between samples. It mainly includes two stages. In stage-I, we use starGAN to train a style transfer model, which generates images of multiple camera styles for increasing the quantity and diversity of samples. In stage-II, we propose incremental optimization learning, which iterates between similarity grouping and CNN model optimization to progressively explore the potential similarities of all training samples. Furthermore, with the purpose of reducing the impact of label noise on performance, we propose a new ranking-guided triplet loss, which is on the basis of similarity and does not require any label to select reliable triple samples. We perform a mass of experiments on Market-1501, and DukeMTMC-reID datasets prove that the proposed CVSE is competitive to the most advanced methods.

Random style transfer for person re-identification with one example

Person re-identification with only one labeled image for each identify can not eliminate the style variations of different cameras in the same dataset. In this paper, we propose a random style transfer strategy that randomly transforms the labeled images on the one-example person re-identification task. In this strategy, we focus on twofolds: 1) Randomly transform the camera style of labeled images and unlabeled images during the training stage and 2) use the average feature of labeled data and its camera style transform data to estimate pseudo label on unlabeled data. Notably, our strategy exhibits state-of-the-art performance on large-scale image datasets and its Rank-1 accuracy outperforms the state-of-the-art method by 10.3% points on Market-1501, and 8.4% points on DukeMTMC-reID.

Improving Domain-Adaptive Person Re-Identification by Dual-Alignment Learning With Camera-Aware Image Generation

Domain adaptation in person re-identification (re-ID) has always been challenging, especially for the lack of supervision information on the target domain. Existing methods generally introduced extra supervision by adversarial learning techniques, then added all the augmented data in the training process to optimize the re-ID model. However, the direct utilization of all the generated data not only increases additional computational cost but also ignores the potential correlation between the origin and generated data. In this article, we propose a novel dual-alignment learning framework (DAL) with camera-aware image generation to efficiently and effectively tackle this issue. Specifically, we propose a camera transfer matching module to generate additional training images with different camera styles, and construct the matching pairs with each containing a origin image and one corresponding camera transferred image. To strengthen the correlation of images for each matching pair, we align the pseudo-labels via clustering algorithm to reduce the pseudo-labels distribution discrepancy between the origin and generated images. Besides, to avoid model degeneration affected by some inaccurate pseudo-labels on unlabelled data, we maximize the mutual information to align the image feature representations of matching pair. The DAL allows us to decrease the camera variance and enhance the discrimination ability of re-ID model. Extensive experiments on three large-scale benchmarks demonstrate the superiority of DAL over state-of-the-art methods.

Leveraging Virtual and Real Person for Unsupervised Person Re-Identification

Person re-identification (re-ID) is a challenging instance retrieval problem, especially when identity annotations are not available for training. Although modern deep re-ID approaches have achieved great improvement, it is still difficult to optimize the deep re-ID model and learn discriminative person representation without annotations in training data. To address this challenge, this study considers the problem of unsupervised person re-ID and introduces a novel approach to solve this problem by leveraging virtual and real data. Our approach includes two components: virtual person generation and training of the deep re-ID model . For virtual person generation, we learn a person generation model and a camera style transfer model using unlabeled real data to generate virtual persons with different poses and camera styles. The virtual data is formed as labeled training data, enabling subsequent training deep re-ID model in supervision. For training of the deep re-ID model, we divide it into three steps: 1) pre-training a coarse re-ID model by using virtual data; 2) collaborative filtering based positive pair mining from the real data; and 3) fine-tuning of the coarse re-ID model by leveraging the mined positive pairs and virtual data. The final re-ID model is achieved by iterating between step 2 and step 3 until convergence. Extensive experiments demonstrate the effectiveness of our method. Experimental results on two large-scale datasets, Market-1501 and DukeMTMC-reID, show the advantages of our method over state-of-the-art approaches in unsupervised person re-ID. Our code is now available online. 1

A New Strategy for One-Example Person re-ID: Exploit the Unlabeled Data Gradually Base on Style-Transferred Images

As a research field of symmetry journals, computer vision has received more and more attention. Person re-identification (re-ID) has become a research hotspot in computer vision. We focus on one-example person re-ID, where each person only has one labeled image in the dataset, and other images are unlabeled. There are two main challenges of the task, the insufficient labeled data, and the lack of labeled images cross-cameras. In dealing with the above issue, we propose a new one-example labeling scheme, which generates style-transferred images by CycleGAN (Cycle Generative Adversarial Networks) to ensure that for each person, there is one labeled image under each camera style. Then a self-learning framework is adopted, which iteratively train a CNN (Convolutional Neural Networks) model with labeled images and labeled style-transferred images, and mine the reliable images to assign a pseudo label. The experimental results prove that by integrating the camera style transferred images, we effectively expand the dataset, and the problem of low recognition rate caused by the lack of labeled pedestrian pictures across cameras is effectively solved. Notably, the rank-1 accuracy of our method outperforms the state-of-the-art method by 8.7 points on the Market-1501 dataset, and 6.3 points on the DukeMTMC-ReID dataset.

Unsupervised Person Re-identification via Cross-camera Similarity Exploration.

Most person re-identification (re-ID) approaches are based on supervised learning, which requires manually annotated data. However, it is not only resource-intensive to acquire identity annotation but also impractical for large-scale data. To relieve this problem, we propose a cross-camera unsupervised approach that makes use of unsupervised style-transferred images to jointly optimize a convolutional neural network (CNN) and the relationship among the individual samples for person re-ID. Our algorithm considers two fundamental facts in the re- ID task, i.e., variance across diverse cameras and similarity within the same identity. In this paper, we propose an iterative framework which overcomes the camera variance and achieves across-camera similarity exploration. Specifically, we apply an unsupervised style transfer model to generate style-transferred training images with different camera styles. Then we iteratively exploit the similarity within the same identity from both the original and the style-transferred data. We start with considering each training image as a different class to initialize the Convolutional Neural Network (CNN) model. Then we measure the similarity and gradually group similar samples into one class, which increases similarity within each identity. We also introduce a diversity regularization term in the clustering to balance the cluster distribution. The experimental results demonstrate that our algorithm is not only superior to state-of-the-art unsupervised re-ID approaches, but also performs favorably compared with other competing unsupervised domain adaptation methods (UDA) and semi-supervised learning methods.

Stargan-based camera style transfer for person retrieval

Person retrieval is also known as person re-identification (ReID) aiming to match person among cross cameras. Although the results of the person ReID have performed well in small datasets, the issues of the large number of identities in real scenarios or with more cameras have not been fully investigated. Being an image retrieval task under cross multi-cameras of intelligent video security, person ReID is influenced by the image style change caused by different camera illumination and view angles. The number of cameras in the latest datasets is increasing and more camera transfer models need to be trained. Traditional methods of generative adversarial network (GAN) can only handle transfer of two domains. To facilitate the research towards solving these problems, we use star generative adversarial networks (StarGAN) to transfer the image from one camera to another camera in the latest large benchmark datasets. We train multiple transfer models simultaneously, minimising the bias among different cameras. Label smooth regularisation (LSR) algorithm is utilised to mitigate the effects of noise in the model. We learn part-based descriptors from pedestrian samples to generate robust feature representation. Our work is competitive compared to the state-of-the-art.

Stargan-based camera style transfer for person retrieval

Person retrieval is also known as person re-identification (ReID) aiming to match person among cross cameras. Although the results of the person ReID have performed well in small datasets, the issues of the large number of identities in real scenarios or with more cameras have not been fully investigated. Being an image retrieval task under cross multi-cameras of intelligent video security, person ReID is influenced by the image style change caused by different camera illumination and view angles. The number of cameras in the latest datasets is increasing and more camera transfer models need to be trained. Traditional methods of generative adversarial network (GAN) can only handle transfer of two domains. To facilitate the research towards solving these problems, we use star generative adversarial networks (StarGAN) to transfer the image from one camera to another camera in the latest large benchmark datasets. We train multiple transfer models simultaneously, minimising the bias among different cameras. Label smooth regularisation (LSR) algorithm is utilised to mitigate the effects of noise in the model. We learn part-based descriptors from pedestrian samples to generate robust feature representation. Our work is competitive compared to the state-of-the-art.

Exploring Latent Information for Unsupervised Person Re-Identification by Discriminative Learning Networks

For unsupervised domain adaption in person re-identification (Re-ID) tasks, the generally used label estimation approaches simply use the global features or the uniform part features. They often neglect the variations of samples having the same identity caused by occlusion, misalignment and uncontrollable camera settings. In this paper, we propose a discriminative learning network with target domain latent information (LatentDLN) to enhance the generalization ability of the Re-ID model. Specifically, to generate a discriminative and robust representation, two types of latent information in the samples from the target domain are explored by the multi-branch deep structure. First, the key points based valid region information is used to leverage the local and global cues in human body, and then a heuristic distance metric learning method based on the global features and the local features is proposed to effectively evaluate the similarity between different images. Second, the camera style transferred images are used as augmentation data to bridge the gap between different cameras in target domains. Moreover, the re-rank mechanism based on reciprocal neighbors is designed to improve the quality of the label estimation. Experimental results on Market-1501, DukeMTMC-ReID and MSMT17 datasets validate the significant effectiveness of the proposed LatentDLN for unsupervised Re-ID.

Cross-domain person re-identification using Dual Generation Learning in camera sensor networks

Cross-domain person re-identification (Re-ID) in camera sensor networks is a challenge task due to large domain-style and camera-style variances. In this work, we propose a novel deep learning method called Dual Generation Learning (DGL) for cross-domain person Re-ID, which simultaneously considers domain and camera styles by expanding training samples. Correspondingly, we design a three-branch deep model with different losses. We further propose Hybrid Triplet Loss (HTL) to deal with the combination of the source dataset, the target dataset and their expansions. Thus, the learned features are robust to domain shifts and camera differences. The experimental results prove that DGL achieves the promising generalization ability and accuracy compared with the state-of-the-art methods.

Domain Adaptive Person Re-Identification via Camera Style Generation and Label Propagation

Unsupervised domain adaptation in person re-identification resorts to labeled source data to promote the model training on target domain, facing the dilemmas caused by large domain shift and large camera variations. The non-overlapping labels challenge that source domain and target domain have entirely different persons further increases the re-identification difficulty. In this paper, we propose a novel algorithm to narrow such domain gaps. We derive a camera style adaptation framework to learn the style-based mappings between different camera views, from the target domain to the source domain, and then we can transfer the identity-based distribution from the source domain to the target domain on the camera level. To overcome the non-overlapping labels challenge and guide the person re-identification model to narrow the gap further, an efficient and effective soft-labeling method is proposed to mine the intrinsic local structure of the target domain through building the connection between GAN-translated source domain and the target domain. Experiment results conducted on real benchmark datasets indicate that our method gets state-of-the-art results.

Intersubjektivitas dan Gaya Kamera dalam Film Etnografi

This article is about the use of observational camera style in articulating intersubjectivity. The case is my experience in making Lukas' Moment (2005), an ethnographic film shot in Merauke, Papua. The film is about Papuans’ intersubjectivity toward indigeneity in the post-2000 riot Papua. I observed the experiences of three Papuan entrepreneurs in processes of building their enterprises which in different ways involve politics of indigeneity. Following the existential-phenomenological anthropology approach, I explicate intersubjectivity as instable formation of subjects engaging in their process of being-in-the-world. In the filming process, I experienced two modalities of observational camera, which I called hunting and fishing modes of visuality. The hunting mode focuses on searching for preconceived images of targeted social scenes to satisfy a research goal, while the fishing mode is about building correspondence with the rhythm of the subject’s social processes in everyday life in order to capture the unexpected social scenes. As it is characterized by establishing cinematic correspondence with the subject’s path of life, I argue that the fishing mode is more suitable as an ethnographic endeavor which aims to broaden the horizon of humanity instead of predicting it.

CamStyle: A Novel Data Augmentation Method for Person Re-Identification.

Person re-identification (re-ID) is a cross-camera retrieval task that suffers from image style variations caused by different cameras. The art implicitly addresses this problem by learning a camera-invariant descriptor subspace. In this paper, we explicitly consider this challenge by introducing camera style (CamStyle). CamStyle can serve as a data augmentation approach that reduces the risk of deep network overfitting and that smooths the CamStyle disparities. Specifically, with a style transfer model, labeled training images can be style transferred to each camera, and along with the original training samples, form the augmented training set. This method, while increasing data diversity against overfitting, also incurs a considerable level of noise. In the effort to alleviate the impact of noise, the label smooth regularization (LSR) is adopted. The vanilla version of our method (without LSR) performs reasonably well on few camera systems in which overfitting often occurs. With LSR, we demonstrate consistent improvement in all systems regardless of the extent of overfitting. We also report competitive accuracy compared with the state of the art on Market-1501 and DukeMTMC-re-ID. Importantly, CamStyle can be employed to the challenging problems of one view learning and unsupervised domain adaptation (UDA) in person re-identification (re-ID), both of which have critical research and application significance. The former only has labeled data in one camera view and the latter only has labeled data in the source domain. Experimental results show that CamStyle significantly improves the performance of the baseline in the two problems. Specially, for UDA, CamStyle achieves state-of-the-art accuracy based on a baseline deep re-ID model on Market-1501 and DukeMTMC-reID. Our code is available at: https://github.com/zhunzhong07/CamStyle .

The Effect of the Handheld Camera on Narration and Ocularization in the Dogme 95 Film

This paper aims, based on the method of case study, to investigate how the handheld camera affects narration and ocularization in the Dogme 95 film. The Dogme 95 (hereafter ‘Dogme’) film is a relatively recent filmmaking movement, which has its own set of strict rules – one of which requires shooting the film with a handheld camera. Regardless of the fact that the utilization of a handheld camera causes specific and recognizable image and movement aesthetics, which can be described as a specific handheld camera style, it is also an indispensable theoretical fact that it very much ‘speaks’ through narration, in particular through focalization and ocularization, which is also theoretically confirmed by theoreticians of the Dogme film. The two most representative, and most prominent films of the Dogme filmmaking movement are Idiots [Idioterne] by Lars von Trier and The Celebration [Festen] by Thomas Vinterberg. The Dogme movement, famous for its inescapable and apparent pseudo-documentary filming style – which enables one to ask the question what is here and how is the reality of reality here – raises additional questions, but also provides answers through thoroughly studying the way in which the handheld camera narrates audiovisually and how narration flows in the two aforementioned films. Additionally, it asks whether there is a difference in narration and ocularization, or if it is disabled, and how the handheld camera becomes the narrator, in symbiosis with the character.
 
 Article received: December 28, 2017; Article accepted: January 10, 2018; Published online: April 15, 2018; Original scholarly paper
 
 How to cite this article: Pandža, Kristian. "The Effect of the Handheld Camera on Narration and Ocularization in the Dogme 95 Film." AM Journal of Art and Media Studies 15 (2018): 83–92. doi: 10.25038/am.v0i15.232

Visual and Experiential Knowledge in Observational Cinema

This article examines observational filmmaking methods, which have been influential in shaping ethnographic cinema. It offers a theoretical discussion of the techniques of observational filmmaking and their exploratory function, providing an analysis of the general features of observational films. The paper examines observational cinema with respect to the degree of control exerted over the filmmaking process, camera style and editing, the role of subjectivity and the interaction of words and images. The article concludes with an analysis of the contribution of observational cinema in the creation of corporeal and experiential forms of knowledge foregrounded in postmodern anthropology.

Reverse Modeling of Auto Part Reconstruction Based on CATIA

Abstract: In the light of the problem of low efficiency in reverse engineering model reconstruction, taking Auto part as object of study, analysis the key technology in this paper. One method is proposed to improve the design efficiency for Auto part model reconstruction. First of all, point cloud is acquired by camera style 3D scanner and processed by Geomagic software; then it is imported to CATIA software for Parameters reproduction and model reconstruction. Results show that this method can improve the efficiency of Auto part model reconstruction, and provide one workable way for CAD model reconstruction with complex surface.

Err, New Boundaries for Fiction?! ER's 'Live' Episode

Err, New Boundaries for Fiction?! ER's 'Live' Episode

Toward a Non-Bourgeois Camera Style

Research Article| December 01 1970 Toward a Non-Bourgeois Camera Style Brian Henderson Brian Henderson Search for other works by this author on: This Site PubMed Google Scholar Film Quarterly (1970) 24 (2): 2–14. https://doi.org/10.2307/1211215 Views Icon Views Article contents Figures & tables Video Audio Supplementary Data Peer Review Share Icon Share Twitter LinkedIn Tools Icon Tools Get Permissions Cite Icon Cite Search Site Citation Brian Henderson; Toward a Non-Bourgeois Camera Style. Film Quarterly 1 December 1970; 24 (2): 2–14. doi: https://doi.org/10.2307/1211215 Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex toolbar search Search Dropdown Menu nav search search input Search input auto suggest search filter All ContentFilm Quarterly Search This content is only available via PDF. Copyright 1971 The Regents of the University of California Article PDF first page preview Close Modal You do not currently have access to this content.