Italy

This paper presents TCE: Temporally Coherent Embeddings for self-supervised video representation learning. The proposed method exploits inherent structure of unlabeled video data to explicitly enforce temporal coherency in the embedding space, rather than indirectly learning it through ranking or predictive proxy tasks. In the same way that high-level visual information in the world changes smoothly, we believe that nearby frames in learned representations will benefit from demonstrating similar properties. Using this assumption, we train our TCE model to encode videos such that adjacent frames exist close to each other and videos are separated from one another. Using TCE we learn robust representations from large quantities of unlabeled video data. We thoroughly analyse and evaluate our self-supervised learned TCE models on a downstream task of video action recognition using multiple challenging benchmarks (Kinetics400, UCF101, HMDB51). With a simple but effective 2D-CNN backbone and only RGB stream inputs, TCE pre-trained representations outperform all previous selfsupervised 2D-CNN and 3D-CNN pre-trained on UCF101. The code and pre-trained models for this paper can be downloaded at: https://github.com/csiro-robotics/TCE

ICPR 2020

Temporally Coherent Embeddings for Self-Supervised Video Representation Learning

video representation learning

temporal learning

spatio-temporal learning

action recognition

representation learning

contrastive learning

self-supervised learning

**About**
 The PErvasive Technologies Related to Assistive Environments (PETRA) conference is a highly interdisciplinary conference that focuses on computational and engineering approaches to improve the quality of life and enhance human performance in a wide range of settings, in the workplace, at home, in public spaces, urban environments, and other. Outcomes of this conference have a broad impact in application areas that include, manufacturing, transportation, healthcare, energy systems, security and safety, robotics, biomedicine, environment and conservation, and many others. PETRA brings together very different types of technologies to also address important social and healthcare issues for sensitive populations, such as the elderly, persons suffering from chronic conditions such as Alzheimer’s, Parkinson’s, Cerebral Palsy, and other disabilities or traumas.

 PETRA (which means “stone” in Greek) addresses important human living needs. From a technological perspective, PETRA motivates and inspires new scientific thrusts, innovations and discoveries to improve living environments, a particularly important issue, in an increasingly crowded and complex world where the need for inclusiveness and connectivity with the rest of the world is key. For the elderly in particular, PETRA addresses the fact that, as people grow older; they will increasingly rely on technology to be able to prolong staying in their homes and enjoy an independent life.

 The PETRA 2021 proceedings will be published in the ACM Digital Library as part of the ACM ICPS program. The ISBN number assigned to PETRA '21 is 978-1-4503-8792-7

PETRA 2021

The PErvasive Technologies Related to Assistive Environments (PETRA) conference is a highly interdisciplinary conference that focuses on computational and engineering approaches to improve the quality of life and enhance human performance in a wide range of settings, in the workplace, at home, in public spaces, urban environments, and other. Outcomes of this conference have a broad impact in application areas that include, manufacturing, transportation, healthcare, energy systems, security and safety, robotics, biomedicine, environment and conservation, and many others. PETRA brings together very different types of technologies to also address important social and healthcare issues for sensitive populations, such as the elderly, persons suffering from chronic conditions such as Alzheimer’s, Parkinson’s, Cerebral Palsy, and other disabilities or traumas.

poster

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. 



This paper studies the problem of Person Re-Identification (ReID)
for large-scale applications. Recent research efforts have been devoted
to building complicated part models, which introduce considerably
high computational cost and memory consumption, inhibiting its practicability
in large-scale applications. This paper aims to develop a novel learning
strategy to find efficient feature embeddings while maintaining the
balance of accuracy and model complexity. More specifically, we find
by enhancing the classical triplet loss together with cross-entropy
loss, our method can explore the hard examples and build a discriminant
feature embedding yet compact enough for large-scale applications.
Our method is carried out progressively using Bayesian optimization,
and we call it the Progressive Learning Algorithm (PLA). Extensive
experiments on three large-scale datasets show that our PLA is comparable
or better than the state-of-the-arts. Especially, on the challenging
Market-1501 dataset, we achieve Rank-1=94.7%/mAP=89.4% while saving
at least 30\% parameters than strong part models.

Explain2Attack: Text Adversarial Attacks via Cross-Domain Interpretability

Neural network pruning with suitable retraining
can yield networks with considerably fewer parameters than the
original with comparable degrees of accuracy. Typical pruning
methods require large, fully trained networks as a starting point
from which they perform a time-intensive iterative pruning
and retraining procedure to regain the original accuracy. We
propose a novel pruning method that prunes a network in real-time
during training, reducing the overall training time to achieve
an efficient compressed network. We introduce an activation
density based analysis to identify the optimal relative sizing
or compression for each layer of the network. Our method is
architecture agnostic, allowing it to be employed on a wide
variety of systems. For VGG-19 and ResNet18 on CIFAR-10,
CIFAR-100, and TinyImageNet, we obtain exceedingly sparse
networks (up to 200× reduction in parameters and over 60×
reduction in inference compute operations in the best case)
with accuracy comparable to the baseline network. By reducing
the network size periodically during training, we achieve
total training times that are shorter than those of previously
proposed pruning methods. Furthermore, training compressed
networks at different epochs with our proposed method yields
considerable reduction in training compute complexity (1.6×
to 3.2× lower) at near iso-accuracy as compared to a baseline
network trained entirely from scratch.

Activation Density driven Efficient Pruning inTraining


This paper studies the problem of Person Re-Identification (ReID)
for large-scale applications. Recent research efforts have been devoted
to building complicated part models, which introduce considerably
high computational cost and memory consumption, inhibiting its practicability
in large-scale applications. This paper aims to develop a novel learning
strategy to find efficient feature embeddings while maintaining the
balance of accuracy and model complexity. More specifically, we find
by enhancing the classical triplet loss together with cross-entropy
loss, our method can explore the hard examples and build a discriminant
feature embedding yet compact enough for large-scale applications.
Our method is carried out progressively using Bayesian optimization,
and we call it the Progressive Learning Algorithm (PLA). Extensive
experiments on three large-scale datasets show that our PLA is comparable
or better than the state-of-the-arts. Especially, on the challenging
Market-1501 dataset, we achieve Rank-1=94.7%/mAP=89.4% while saving
at least 30\% parameters than strong part models.

Contextual Classification Using Self-Supervised Auxiliary Models for Deep Neural Networks


Scene text detection (STD) aims to locate text in
images and plays an important role in many computer vision
tasks including automatic driving and text recognition systems.
Recently, deep neural networks (DNNs) have been widely and
successfully used in scene text detection, leading to plenty
of DNN-based STD methods including regression-based and
segmentation-based STD methods. However, recent studies have
also shown that DNN is vulnerable to adversarial attacks, which
can significantly degrade the performance of DNN models. In
this paper, we investigate the robustness of DNN-based STD
methods against adversarial attacks. To this end, we propose
a generic and efficient attack method to generate adversarial
examples, which are produced by adding small but imperceptible
adversarial perturbation to the input images. Experiments on
attacking four various models and a real-world STD engine of
Google optical character recognition (OCR) show that the stateof-the-art DNN-based STD methods including regression-based
and segmentation-based methods are vulnerable to adversarial
attacks.

Transferable Adversarial Attacks for Deep SceneText Detection

It is well-known that identity-unrelated variations (e.g., viewpoint or illumination) degrade the performances of face recognition methods. In order to handle this challenge, a robust method for disentangling the identity and view representations has drawn an attention in the machine learning area. However, existing methods learn discriminative features which require a manual supervision of such factors of variations. In this paper, we propose a novel disentangling framework through modeling three representations of identity, viewpoint, and residues (i.e., identity and pose unrelated) which do not require supervision of the variations. By jointly modeling the three representations, we enhance the disentanglement of each representation and achieve robust face recognition performance. Further, the learned viewpoint representation can be utilized for pose estimation or editing of a posed facial image. Extensive quantitative and qualitative evaluations verify the effectiveness of our proposed method which disentangles identity, viewpoint, and residues of facial images.

Unsupervised Disentangling of Viewpoint and Residues Variations by Substituting Representations for Robust Face Recognition

Object detection has long been an important issue in the discipline of scene understanding. Existing researches mainly focus on the object itself, ignoring its surrounding environment. In fact, the surrounding environment provides abundant information to help detectors classify and locate objects. This paper proposes CRPDet, viz. CenterRepp Detector, a framework for object detection. The main function of CRPDet is accomplished by the CenterRepp module, which takes into account the surrounding environment by predicting the distribution of the central representative points. CenterRepp converts labeled object frames into the mean and standard variance of the sampling points’ distribution. This helps increase the receptive field of objects, breaking the limitation of object frames. CenterRepp defines a position-fixed center point with significant weights, avoiding to sample all points in the surroundings. Experiments on the COCO test-dev detection benchmark demonstrates that our proposed CRPDet has comparable performance with state-of-the-art detectors, achieving 39.4 mAP with 51 FPS tested under single size input.

CenterRepp: Predict Central Representative Point Set’s Distribution For Detection


Reconstruction of 3D human bodies, from images, faces many challenges, due to it generally being an ill-posed problem. In this paper we present a method to reconstruct 3D human bodies from multi-view images, through learning an
implicit function to represent 3D shape, based on multi-scale features extracted by multi-stage end-to-end neural networks. Our model consists of several end-to-end hourglass networks for extracting multi-scale features from multi-view images, and a
fully connected network for implicit function classification from these features. Given a 3D point, it is projected to multi-view images and these images are fed into our model to extract multi-scale features. The scales of features extracted by the hourglass networks decrease with the depth of our model, which represents
the information from local to global scale. Then, the multi-scale features as well as the depth of the 3D point are combined to a new feature vector and the fully connected network classifies the feature vector, in order to predict if the point lies inside or outside of the 3D mesh. The advantage of our method is that we use both local and global features in the fully connected network and represent the 3D mesh by an implicit function, which is more memory-efficient. Experiments on public datasets demonstrate that our method surpasses previous approaches in terms of the accuracy of 3D reconstruction of human bodies from images.

Learning to Implicitly Represent 3D Human Body From Multi-scale Features and Multi-view Images


Ghost-like artifact caused by ill-exposed and motion areas is one of the most challenging problems in high dynamic range (HDR) image reconstruction. When the motion range is small, previous methods based on optical flow or patch-match can suppress ghost-like artifacts by first aligning input images before merging them. However, they are not robust enough and still produce artifacts for challenging scenes where large foreground motions exist. To this end, we propose a deep network with an attention module and motion-emphasized loss function to produce ghost-free HDR images. In the attention module, we use the channel and spatial attention to guide the network to emphasize important components such as motion and saturated areas automatically. To be robust to images with different resolutions and objects with distinct scales, we adopt the selective kernel network as the basic framework for channel attention. In addition to the attention module, the motion-emphasized loss function based on the motion and ill-exposed areas mask is designed to help the network reconstruct motion areas. Experiments on the public dataset indicate that the proposed SK-AHDRNet produces ghost-free results where detail in ill-exposed areas is well recovered. The proposed method scores 43.17 with PSNR metric and 61.02 with HDR-VDP-2 metric on test which outperforms all conventional works. According to quantitative and qualitative evaluations, the proposed method can achieve state-of-the-art performance.

Selective Kernel and Motion-emphasized Loss Based Attention-guided Network for HDR Imaging of Dynamic Scenes


Video anomaly detection is a challenging task not only because it involves solving many sub-tasks such as motion representation, object localization and action recognition, but also because it is commonly considered as an unsupervised learning problem that involves detecting outliers. Traditionally, solutions to this task have focused on the mapping between video frames and their low-dimensional features, while ignoring the spatial connections of those features. Recent solutions focus on analyzing these spatial connections by using hard clustering techniques, such as K-Means, or applying neural networks to map latent features to a general understanding, such as action attributes. In order to solve video anomaly in the latent feature space, we propose a deep probabilistic model to transfer this task into a density estimation problem where latent manifolds are generated by a deep denoising autoencoder and clustered by expectation maximization. Evaluations on several benchmarks datasets show the strengths of our model, achieving outstanding performance on challenging datasets.

Video Anomaly Detection by Estimating Likelihood of Representations

Human beings not only have the ability to recognize novel unseen classes, but also can incrementally incorporate the new classes to existing knowledge preserved. However, zero- shot learning models assume that all seen classes should be known beforehand, while incremental learning models cannot recognize unseen classes. This paper introduces a novel and challenging task of Incrementally Zero-Shot Detection (IZSD), a practical strategy for both zero-shot learning and class- incremental learning in real-world object detection. An innovative end-to-end model – IZSD-EVer was proposed to tackle this task that requires incrementally detecting new classes and detecting the classes that have never been seen. Specifically, we propose a novel extreme value analyzer to detect objects from old seen, new seen, and unseen classes, simultaneously. Additionally and technically, we propose two innovative losses, i.e., background- foreground mean squared error loss alleviating the extreme imbalance of the background and foreground of images, and projection distance loss aligning the visual space and semantic spaces of old seen classes. Experiments demonstrate the efficacy of our model in detecting objects from both the seen and unseen classes, outperforming the alternative models on Pascal VOC and MSCOCO datasets.

Temporally Coherent Embeddings for Self-Supervised Video Representation Learning

Downloads

Next from ICPR 2020

Explain2Attack: Text Adversarial Attacks via Cross-Domain Interpretability

Similar lecture

Automated system to measure Tandem Gait to assess executive functions in children

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES