Italy

Low-light is a challenging environment for both human and computer vision to perform tasks such as object classification and detection. Recent works have shown potential in employing enhancements algorithms to support and improve such tasks in low-light, however there has not been any focused analysis to understand the direct effects that low-light enhancement have on an object detector. This work aims to quantify and visualize such effects on the multi-level abstractions involved in network-based object detection. First, low-light image enhancement algorithms are employed to enhance real low-light images, and then followed by deploying an object detection network on the low-light as well as the enhanced counterparts. A comparison of the activations in different layers, representing the detection features, are used to generate statistics in order to quantify the enhancements&#39; contribution to detection. Finally, this framework was used to analyze several low-light image enhancement algorithms and identify their impact on the detection
model and task. This framework can also be easily generalized to any convolutional neural network-based models for the analysis of different enhancements algorithms and tasks.

ICPR 2020

Exploring the Contributions of Low-light Image Enhancement to Network-based Object Detection

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. 


Low-light is a challenging environment for both human and computer vision to perform tasks such as object classification and detection. Recent works have shown potential in employing enhancements algorithms to support and improve such tasks in low-light, however there has not been any focused analysis to understand the direct effects that low-light enhancement have on an object detector. This work aims to quantify and visualize such effects on the multi-level abstractions involved in network-based object detection. First, low-light image enhancement algorithms are employed to enhance real low-light images, and then followed by deploying an object detection network on the low-light as well as the enhanced counterparts. A comparison of the activations in different layers, representing the detection features, are used to generate statistics in order to quantify the enhancements' contribution to detection. Finally, this framework was used to analyze several low-light image enhancement algorithms and identify their impact on the detection
model and task. This framework can also be easily generalized to any convolutional neural network-based models for the analysis of different enhancements algorithms and tasks.

technical paper

This paper presents a deep learning-based method
to estimate gait attributes (e.g., stately, cool, relax, etc.). Similarly
to the existing studies on relative attribute, human perception-
based annotations on the gait attributes are given to pairs of
gait videos (i.e., the ﬁrst one is better, tie, and the second one
is better), and the relative annotations are utilized to train a
ranking model of the gait attribute. More speciﬁcally, we design
a Siamese (i.e., two-stream) network which takes a pair of
gait inputs and output gait attribute score for each. We then
introduce a suitable loss function called a signed contrastive loss
to train the network parameters with the relative annotation.
Unlike the existing loss functions for learning to rank does
not inherit a nice property of a quadratic contrastive loss, the
proposed signed quadratic contrastive loss function inherits the
nice property. The quantitative evaluation results reveal that
the proposed method shows better or comparable accuracies of
relative attribute prediction against the baseline methods.

Deep Gait Relative Attribute using a Signed Quadratic Contrastive Loss

Generating as-is 3D Models is constantly explored for various construction management applications. The industry has been dependent on either manual or semi-automated workflows for the Scan-to-BIM process, which is laborious as well as time taking. Recently machine learning has opened avenues to recognize geometrical elements from point clouds but has not been much used because of the insufficient labeled dataset. This study aims to set up a semi-automated workflow to create labeled data sets which can be used to train ML algorithms for element identification purpose. The study proposes an interactive user interface using a gaming engine within a mixed reality environment. A workflow for fusing as-is spatial information with the AR/VR based information is presented in Unity 3D. A user friendly UI is then developed and integrated with the VR environment to help the user to choose the category of the component by visualization. This results in the generation of an accurate as-is 3D Model, which does not require much computation or time. The intention is to propose a smooth workflow to generate datasets for learning-based methodologies in a streamlined Scan-to-BIM Process. However, this process requires user domain knowledge and input. The dataset can be continuously increased and improved to get automated results later.


Mixed Reality-Based Dataset Generation for Learning Based Scan-to-BIM

Remote working and collaboration is important towards helping workplaces to become flexible and productive. The significance of remote working has also been highlighted in the COVID-19 pandemic period in which mobility restrictions were enforced. This paper presents the development of an augmented reality platform, aiming to assist workers in remote collaboration and training. The platform consists of two communicating apps intended to be used by a remote supervisor (located e.g., at home) and an on-site worker, and uses intuitive digital annotations that enrich the physical environment of the workplace, thereby facilitating the execution of on-site tasks. The proposed platform was used and evaluated in user trials, demonstrating its usefulness and virtue by assessing its performance, worker satisfaction and task completion time.

An Augmented Reality-based Remote Collaboration Platform for Worker Assistance

The future civilian, and professional autonomous vehicles to be realised into the market should apprehend and interpret the road in a manner similar to the human drivers. In structured urban environments where signs, road lanes and markers are well defined and ordered, landmark-based road tracking and localisation has significantly progressed during the last decade with many autonomous vehicles to make their debut into the market. However, in semi-structured and rural environments where traffic infrastructure is deficient, the autonomous driving is hindered by significant challenges. The paper at hand presents a Lidar-based method for road boundaries detection suitable for a service robot operation in rural and semi-structured environments. Organised Lidar data undergo a spatial distribution processing method to isolate road limits in a forward looking horizon ahead of the robot. Stereo SLAM is performed to register subsequent road limits and RANSAC is applied to identify edges that correspond to road segments. In addition, the robot traversable path is estimated and progressively merged with Bézier curves to create a smooth trajectory that respects vehicle kinematics. Experiments have been conducted on data collected from our robot on a semi-structured urban environment, while the method has also been evaluated on KITTI dataset exhibiting remarkable performance.

PRRoad tracking in semi-structured environments using spatial distribution of Lidar data

Breathing rate together with heart rate, blood pressure,
and body temperature, are considered as essential and vital
signs of human body. In this demo, we present a novel system to
estimate the respiratory rate using a low-cost, non-intrusive, and
passive infrared thermopile sensor array. The thermopile sensor
array can detect the thermal infrared energy and generate a
low-resolution video from subjects within its viewing range. We
collected low resolution 8 8-pixel thermal video clips from 40
subjects at resting position from a distance of one-meter between
the thermopile sensor array and the subjects. We processed
the image frames using Haar-like filters and generated a one-dimensional
feature signal from each video. The signal exhibits
an almost periodic behavior following the respiratory movements.
A peak detection algorithm is used to detect each breath. The
proposed method is computationally efficient because it is based
on multiplierless Haar-type feature extraction and peak detection.
We achieved a high positive correlation rate of 97.77% between
the actual respiratory rate that was measured by monitoring
the chest movements during inhalation and exhalation and the
estimated rate in our dataset. We will demonstrate this IR array
based respiration rate detection system during the demo session.

Respiratory rate estimation using a low-resolution infrared sensor array

Feathers are fascinating natural components of birds and other, extinct dinosaurs. They exhibit a variety of crucial biological properties that are studied in multiple fields. Here we present work in visual analysis of feathers that leverages image-processing methods to observe important physical data about the structure and shape of feathers. Such data may be used to inform ornithology or paleontology as well as used to synthesize accurate computerized representations of feathers for high-fidelity visualization and interactive rendering.

Computer-Aided Visual Analysis of Feathers

This paper proposes a system for pose estimation on monkeys, in diverse and challenging scenarios, and under complex social interactions by using OpenPose. In comparison to most animals used for research, Monkeys present additional difficulties for pose estimation. Multiple degrees of freedom, unique complex postures, intricated social interactions, among others. Our monkey OpenPose trained model is robust against these difficulties. It achieves similar performance as in human pose estimation models, and it can run in real-time.

Multiple Monkey Pose Estimation Using OpenPose

This work describe an algorithm for the automatic analysis of the waggle dance of honeybees. The algorithm analyses a video of a beehive with 13,624 frames, acquired at 25 frames/second. The algorithm employs the following traditional image processing steps: conversion to grayscale, low pass filtering, background subtraction, thresholding, tracking and clustering to detect run of bees that perform waggle dances. The algorithm detected 44,530 waggle events, i.e. one bee waggling in one time frame, which were then clustered into 511 waggle runs. Most of these were concentrated in one section of the hive. The accuracy of the tracking was 90% and a series of metrics like intra-dance variation in angle and duration were found to be consistent with literature. Whilst this algorithm was tested on a single video, the
ideas and steps, which are simple as compared with Machine and Deep Learning techniques, should be attractive for researchers in this field who are not specialists in more complex techniques.

Automatic Analysis of Bees’ Waggle Dance

We investigate the task of entity counting in overhead imagery from the perspective of re-purposing representations learned from ground imagery, e.g., ImageNet, via feature adaptation. We explore two directions of feature adaptation and analyze their performances using two popular aerial datasets for vehicle counting: PUCPR+ and CARPK. First, we explore proxy self-supervision tasks such as RotNet, jigsaw, and image inpainting to re-fine the pretrained representation. Second, we insert additional network layers to adaptively select suitable features (e.g., squeeze and excitation blocks) or impose desired properties (e.g., using active rotating filters for rotation invariance). Our experimental results show that different adaptations produce different amounts of performance improvements depending on data characteristics.

Fine-tuning for one-look regression vehicle counting in low-shot aerial datasets

3D city-scale point cloud stitching is a critical component for large data collection, environment change detection, in which massive amounts of 3D data are captured under different times and conditions. 
This paper proposes a novel point cloud stitching approach, that automatically and accurately stitches multiple city-scale point clouds, which only share relatively small overlapping areas, into one single model for a larger geographical coverage. The proposed method firstly employs 2D image mosaicking techniques to estimate 3D overlapping areas among multiple point clouds, then applies 3D point cloud registration techniques to estimate the most accurate transformation matrix for 3D stitching.
The proposed method is quantitatively evaluated on city-scale reconstructed point cloud dataset and real-world city LiDAR dataset, in which, our method outperforms other competing methods with significant margins and achieved the highest precision score, recall score, and F-score.
Our method makes an important step towards automatic and accurate city-scale point cloud data stitching, which could be used in a variety of applications.

Premium content

Downloads

Next from ICPR 2020

Deep Gait Relative Attribute using a Signed Quadratic Contrastive Loss

Similar lecture

Dynamic Low-Light Image Enhancement for Object Detection via End-to-End Training