Italy

It is widely shown that Convolutional Neural Networks (CNNs) are vulnerable to adversarial examples on most recognition tasks, such as image classiﬁcation and segmentation. However, few work studies the more complicated task – 3D human pose estimation. This task often requires large-scale datasets, specialized network architectures, and it can be solved either from single-view RGB images or from multi-view RGB images. In this paper, we make the ﬁrst attempt to investigate the robustness of current state-of-the-art 3D human pose estimation methods. To this end, we build four representative baseline models, where most of the current methods can be generally classiﬁed as one of them. Furthermore, we design targeted adversarial attacks to detect whether 3D pose estimators are robust to different camera parameters. For different types of methods, we present a comprehensive study of their robustness on the large-scale Human3.6M benchmark. Our work shows that different methods vary signiﬁcantly in their resistance to adversarial attacks. Through extensive experiments, we show that multi-view 3D pose estimators can be more vulnerable to adversarial examples. We believe that our efforts can shed light on future works to design more robust 3D human pose estimators.

ICPR 2020

On the Robustness of 3D Human Pose Estimation

multi-view systems

3d human pose estimation

adversarial attacks

ICPR 2020 is the premier international conference on Pattern Recognition, gathering leading researchers from around the world to present their latest achievements in Pattern Analysis and Recognition, applied Artificial Intelligence and Machine Learning. The topic of the year is " Putting Artificial intelligence to work on Patterns". 
For the first time, ICPR has employed a two-round review process similar to journal submissions, with new papers allowed to be submitted in either the first or second round and papers submitted in the first round and not accepted allowed to be revised and re-submitted for second round review. 

In this edition of the **ICPR 2020** Conference, there are 1407 Oral/Poster presentations organized in five Tracks: 
- Artificial Intelligence, Machine Learning for Pattern Analysis; 
- Biometrics, Human Analysis and Behavior Understanding; Computer Vision, Robotics and Intelligent Systems; 
- Document and Media Analysis; and 
- Image and Signal Processing. 

The **ICPR 2020** Main Conference program also includes 7 Keynotes from prominent world-renowned scientists, 8 Tutorials, 28 Demos, 8 Challenges, 2 Panels, as well as an Industry/Sponsor Exhibition. In addition, 41 Workshops on a range of important current and emerging research topics complete the ICPR 2020 program. 


poster

3D human pose estimation (HPE) is crucial in many fields, such as human behavior analysis, augmented reality/virtual reality (AR/VR) applications, and self-driving industry. Videos that contain multiple potentially occluded people captured from freely moving monocular cameras are very common in real-world scenarios, while 3D HPE for such scenarios is quite challenging, partially because there is a lack of such data with accurate 3D ground truth labels in existing datasets. In this paper, we propose a temporal regression network with a gated convolution module to transform 2D joints to 3D and recover the missing occluded joints in the meantime. A simple yet effective localization approach is further conducted to transform the normalized pose to the global trajectory. To verify the effectiveness of our approach, we also collect a new moving camera multi-human (MMHuman) dataset that includes multiple people with heavy occlusion captured by moving cameras. The 3D ground truth joints are provided by accurate motion capture (MoCap) system. From the experiments on static-camera based Human3.6M data and our own collected moving-camera based data, we show that our proposed method outperforms most state-of-the-art 2D-to-3D pose estimation methods, especially for the scenarios with heavy occlusions.

Exploring Severe Occlusion: Multi-Person 3D Pose Estimation with Gated Convolution

Face recognition is a widely accepted biometric verification tool, as the face contains a lot of information about the identity of a person. In this study, a 2-step neural-based pipeline is presented for matching 3D facial shape to multiple DNA-related properties (sex, age, BMI and genomic background). The first step consists of a triplet loss-based metric learner that compresses facial shape into a lower dimensional embedding while preserving information about the property of interest. Most studies in the field of metric learning have only focused on 2D Euclidean data. In this work, geometric deep learning is employed to learn directly from 3D facial meshes. To this end, spiral convolutions are used along with a novel mesh-sampling scheme that retains uniformly sampled 3D points at different levels of resolution. The second step is a multi-biometric fusion by a fully connected neural network. The network takes an ensemble of embeddings and property labels as input and returns genuine and imposter scores. Since embeddings are accepted as an input, there is no need to train classifiers for the different properties and available data can be used more efficiently. Results obtained by a 10-fold cross-validation for biometric verification show that combining multiple properties leads to stronger biometric systems. Furthermore, the proposed neural-based pipeline outperforms a linear baseline, which consists of principal component analysis, followed by classification with linear support vector machines and a Naïve Bayes-based score-fuser.

3D Facial Matching by Spiral Convolutional Metric Learning and a Biometric Fusion-Net of Demographic Properties

In this paper, we study the task of facial expression recognition under strong occlusion. We are particularly interested in cases where 50\% of the face is occluded, e.g. when the subject wears a Virtual Reality (VR) headset. While previous studies show that pre-training convolutional neural networks (CNNs) on fully-visible (non-occluded) faces improves the accuracy, we propose to employ knowledge distillation to achieve further improvements. First of all, we employ the classic teacher-student training strategy, in which the teacher is a CNN trained on fully-visible faces and the student is a CNN trained on occluded faces. Second of all, we propose a new approach for knowledge distillation based on triplet loss. During training, the goal is to reduce the distance between an anchor embedding, produced by a student CNN that takes occluded faces as input, and a positive embedding (from the same class as the anchor), produced by a teacher CNN trained on fully-visible faces, so that it becomes smaller than the distance between the anchor and a negative embedding (from a different class than the anchor), produced by the student CNN. Third of all, we propose to combine the distilled embeddings obtained through the classic teacher-student strategy and our novel teacher-student strategy based on triplet loss into a single embedding vector.

We conduct experiments on two benchmarks, FER+ and AffectNet, with two CNN architectures, VGG-f and VGG-face, showing that knowledge distillation can bring significant improvements over the state-of-the-art methods designed for occluded faces in the VR setting. Furthermore, we obtain accuracy rates that are quite close to the state-of-the-art models that take as input fully-visible faces. For example, on the FER+ data set, our VGG-face based on concatenated distilled embeddings attains an accuracy rate of 82.75\% on lower-half-visible faces, which is only 2.24\% below the accuracy rate of a state-of-the-art VGG-13 that is evaluated on fully-visible faces. Given that our model sees only the lower-half of the face, we consider this to be a remarkable achievement. In conclusion, we consider that our distilled CNN models can provide useful feedback for the task of recognizing the facial expressions of a person wearing a VR headset.

Teacher-Student Training and Triplet Loss for Facial Expression Recognition under Occlusion

Human body part segmentation refers to the task of predicting the semantic segmentation mask for each body part. Fully supervised body part segmentation methods achieve good performances but require an enormous amount of effort to annotate part masks for training. In contrast to high annotation costs needed for a limited number of part mask annotations, a large number of weak labels such as poses and full body masks already exist and contain relevant information. Motivated by the possibility of using existing weak labels, we propose the first weakly supervised body part segmentation framework. The core idea is first converting the sparse weak labels such as keypoints to the initial estimate of body part masks, and then iteratively refine the part mask predictions. We name the initial part masks estimated from poses the "part priors." With sufficient extra weak labels, our weakly supervised framework achieves a comparable performance (62.0% mIoU) to the fully supervised method (63.6% mIoU) on the Pascal-Person-Part dataset. Furthermore, in the extended semi-supervised setting, the proposed framework outperforms the state-of-art methods. Moreover, we extend our proposed framework to other keypoint-supervised part segmentation tasks such as face parsing.

Weakly Supervised Body Part Segmentation with Pose based Part Priors

We present a novel unsupervised learning approach to image landmark discovery by incorporating the inter-subject landmark consistencies on facial images. This is achieved via an inter-subject mapping module that transforms original subject landmarks based on an auxiliary subject-related structure. To recover from the transformed images back to the original subject, the landmark detector is forced to learn spatial locations that contain the consistent semantic meanings both for the paired intra-subject images and between the paired inter-subject images. Our proposed method is extensively evaluated on two public facial image datasets (MAFL, AFLW) with various settings. Experimental results indicate that our method can extract the consistent landmarks for both datasets and achieve better performances compared to the previous state-of-the-art methods quantitatively and qualitatively.

Unsupervised Learning of Landmarks based on Inter-Intra Subject Consistencies

Most existing works regarding facial demographic estimation are focused on still image datasets, although nowadays the need to analyze video content in real applications is increasing. We propose to tackle gender, age and ethnicity estimation in the context of video scenarios. Our main contribution is to use an attribute-specific quality assessment procedure to select most relevant frames from a video sequence for each of the three demographic modalities. Selected frames are classified with fine-tuned MobileNet models and a final video prediction is obtained with a majority voting strategy. Our validation on three different datasets and our comparison with state-of-the art models, show the effectiveness of the proposed demographic classifiers and the quality pipeline, which allows to reduce both: the number of frames to be classified and the processing time in practical applications; and improves the soft biometrics prediction accuracy.

Attribute-based quality assessment for demographic estimation in face videos

Significant advances have been achieved in face recognition in the last decade thanks to the development of deep learning methods. However, recognizing faces captured in uncontrolled environments is still a challenging problem for the scientific community. In these scenarios, the performance of most of existing deep learning based methods abruptly falls, due to the bad quality of the face images. In this work, we propose to use an activation map to represent the quality information in a face image. Different face regions are analyzed to determine their quality and then only those regions with good quality are used to perform the recognition using a given deep face model. For experimental evaluation, in order to simulate unconstrained environments, three challenging databases, with different variations in appearance, were selected: the Labeled Faces in the Wild Database, the Celebrities in Frontal-Profile in the Wild Database, and the AR Database. Three deep face models were used to evaluate the proposal on these databases and in all cases, the use of the proposed activation map allows the improvement of the recognition rates obtained by the original models in a range from 0.3 up to 31%. The obtained results experimentally demonstrated that the proposal is able to select those face areas with higher discriminative power and enough identifying information,

Quality-based Representation for Unconstrained Face Recognition

The ability to detect, localize and classify objects that are anomalies is a challenging task in the computer vision community. In this paper, we tackle these tasks developing a framework to automatically inspect the railway during the night. Specifically, it is able to predict the presence, the image coordinates and the class of obstacles. To deal with the lowlight environment, the framework is based on thermal images and consists of three different modules that address the problem of detecting anomalies, predicting their image coordinates and classifying them. Moreover, due to the absolute lack of publiclyreleased datasets collected in the railway context for anomaly detection, we introduce a new multi-modal dataset, acquired from a rail drone, used to evaluate the proposed framework. 
Experimental results confirm the accuracy of the framework and its suitability, in terms of computational load, performance, and inference time, to be implemented on a self-powered inspection system.

Anomaly Detection, Localization and Classification for Railway Inspection

Determining the links between large amounts of video data with no prior knowledge of the camera positions is a hard task to automate. From a collection of videos acquired from static cameras simultaneously, we propose a method for finding groups of videos with overlapping fields of view. Each video is first processed individually: at regular time steps, objects are detected and are assigned a category and an appearance descriptor. Next, the video is split into cells at different resolutions and we assign to each cell its story: it consists of the list of objects detected in the cell over time. Once the stories are established for each video, the links between cells of different videos are determined by comparing their stories: two cells are linked if they show simultaneous detections of objects of the same category with similar appearances. Pairs of videos with overlapping fields of view are identified using these links between cells. A link graph is finally returned, in which each node represents a video, and the edges indicate pairs of overlapping videos.
The approach is evaluated on a set of 63 real videos from both public datasets and live surveillance videos and shows promising results.

Story comparison for estimating field of view overlap in a video collection

Recognizing an activity with a single reference sample using metric learning approaches is a promising research field. The majority of few-shot methods focus on object recognition or face-identification. We propose a metric learning approach to reduce the action recognition problem to a nearest neighbor search in embedding space. We encode signals into images and extract features using a deep residual CNN. Using triplet loss, we learn a feature embedding. The resulting encoder transforms features into an embedding space in which closer distances encode similar actions while higher distances encode different actions. Our approach is based on a signal level formulation and remains flexible across a variety of modalities. It further outperforms the baseline on the large scale NTU RGB+D 120 dataset for the One-Shot action recognition protocol by 5.6%. With just 60% of the training data, our approach still outperforms the baseline approach by 3.7%. With 40% of the training data, our approach performs comparably well to the second follow up. Further, we show that our approach generalizes well in experiments on the UTD-MHAD dataset for inertial, skeleton and fused data and the Simitate dataset for motion capturing data. Furthermore, our inter-joint and inter-sensor experiments suggest good capabilities on previously unseen setups.

Premium content

On the Robustness of 3D Human Pose Estimation

Downloads

Next from ICPR 2020

Exploring Severe Occlusion: Multi-Person 3D Pose Estimation with Gated Convolution

Similar lecture

On the Robustness of 3D Human Pose Estimation