Showing papers in "Image and Vision Computing in 2018"

TL;DR: This paper proposes a multi-scale strategy for speeding up marker detection in video sequences by wisely selecting the most appropriate scale for detection, identification and corner estimation.

TL;DR: This work proposes a normalized eigenrepresentation of the class manifold that encodes the targets with minimal information loss, improving the accuracy of random projections encoding while enjoying the same convergence rates.

TL;DR: A critical comparison of different state-of-the-art computer vision methods proposed by researchers for classifying fruit and vegetable is presented.

TL;DR: By exploring contextual information and avoiding errors caused by segmentation, this method performs better than conventional methods and achieves state-of-the-art recognition accuracy.

TL;DR: It is found that DL architectures perform similarly to traditional ones for simpler tasks but report significant improvements in more complex tasks, such as word or sentence recognition, with up to 40% improvement in word recognition rates.

TL;DR: A face PAD solution of attack-specific countermeasures based solely on color texture analysis is proposed and investigated to see how well it generalizes under display and print attacks in different conditions.

TL;DR: A simple method to find out which parts of the human face are more important to achieve a high recognition rate, and use that information during training to force a convolutional neural network to learn discriminative features from all the face regions more equally, including those that typical approaches tend to pay less attention to is proposed.

TL;DR: This paper is an attempt to discuss predominant approaches, its constraints and ways to deal in AIA, and presents performance evaluation measures with relevant and influential image annotation database.

TL;DR: This work focuses on a typical urban human-scene where it aims at predicting an agent's behavior using a stochastic model, fuse the various factors that would contribute to a human motion in different contexts and provides a statistical smooth prediction towards the most likely areas.

TL;DR: In this paper, the authors provide an up-to-date, complete list of available annotated datasets and an in-depth analysis of geometric, hand-crafted, and learned facial representations that are used for facial aging and kinship characterization.

TL;DR: This work proposes a method, Deep Recurrent 3D FAce Reconstruction (DRFAR), to solve the task of multi-view 3D face reconstruction using a subspace representation of the 3D facial shape and a deep recurrent neural network that consists of both a deep convolutional neural network (DCNN) and a recurrent Neural Network (RNN).

TL;DR: This paper proposes a novel embedding method termed focus ranking that can be easily unified into a CNN for jointly learning image representations and metrics in the context of fine-grained fabric image retrieval and shows the superiority of the proposed model over existing metric embedding models.

TL;DR: A CNN-based pedestrian attribute-assisted person re-identification framework that performs the attribute learning by a part-specific CNN to model attribute patterns related to different body parts and fuse them with low-level robust Local Maximal Occurrence features to address the problem of the large variation of visual appearance and location of attributes.

TL;DR: A benchmark database for automatic visual classification methods to evaluate their ability for distinguishing visually similar categories of aquatic macroinvertebrate taxa, and presents the classification results of Convolutional Neural Networks that are widely used for deep learning tasks in large databases.

TL;DR: A surprising result is shown, that perhaps the simplest method of template adaptation, combining deep convolutional network features with template specific linear SVMs, outperforms the state-of-the-art by a wide margin.

TL;DR: This work tackles the problem of hair analysis from unconstrained view by relying only on textures, without a-priori information on head shape and location, nor using body-part classifiers, and achieves segmentation accuracy superior to known state-of-the-art.

TL;DR: This critical review focuses on emerging passive and unobtrusive video-based eye-gaze tracking methods in recent literature, with the aim to identify different research avenues that are being followed in response to the challenges of pervasive eye- gaze tracking.

TL;DR: What makes negative results important, how they should be disseminated and incentivized, and what lessons can be learned from cognitive vision research in this regard are addressed.

TL;DR: The experiments demonstrate that the proposed spatio-temporal modeling of human-object interaction videos for online and off-line recognition is effective and discriminative for human object interaction classification as demonstrated here.

TL;DR: Zhang et al. as mentioned in this paper proposed a general learning framework for jointly estimating human gender and age by attempting to formulate such semantic relationships as a form of near-orthogonality regularization and then to incorporate it into the objective of the joint learning framework.

TL;DR: The main factors affecting the gait features that can be acquired from a 2D video sequence are discussed, proposing a taxonomy to classify them across four dimensions and the results highlight the advantages of using rectified shadow silhouettes over body silhouettes under certain conditions.

TL;DR: An overall vision of the main contributions available so far in the field of context-aware biometric systems and methods is provided, along with a comparison of their features, aims and performances.

TL;DR: In this article, a multi-view dynamic facial action unit detection approach is proposed to detect the presence or absence of a specific action unit in a still image of a human face.

TL;DR: Simple methods to calibrate non-overlapping cameras using markers on the cameras are proposed, which works stably and uses fewer images.

TL;DR: A ternary variational level set model involving L0 gradient regularizer and L0 function regularizer in discrete framework following the Chan-Vese model for image segmentation is proposed and has good performance for segmentation of images with severe noise, outliers or low contrast.

TL;DR: A novel human action descriptor based on skeleton data provided by RGB-D cameras for fast action recognition is proposed, built by interpolating the kinematics of skeleton joints using a cubic spline algorithm.

TL;DR: This work proposes three novel deep learning architectures, which are able to perform a joint detection and pose estimation, where the two tasks gradually decouple, and investigates whether the pose estimation problem should be solved as a classification or regression problem, being this still an open question.

TL;DR: This work proposes a new cascaded regressor for eye center detection that achieves state-of-the-art performance on the BioID, GI4E, and the TalkingFace datasets and improves the robustness of localization by using both advanced features and powerful regression machinery.

TL;DR: A new compact-aware minimum barrier distance for superpixel segmentation (MBS), and a propagation scheme for the cluster centers between adjacent levels on a hierarchical architecture are introduced.

TL;DR: This work proposes informative motion descriptors based on accelerometric data, skeleton joints features and depth maps, and demonstrates their potential to model the motion dynamics, and shows that fusing data from multiple modalities permits better recognition accuracy.