Italy

Recent advances in generalized image understanding have seen a surge in the use of deep convolutional neural networks (CNN) across a broad range of image-based detection, classification and prediction tasks. Whilst the reported performance of these approaches is impressive, this study investigates the hitherto unapproached question of the impact of commonplace image and video compression techniques on the performance of such deep learning architectures. Focusing on the JPEG and H.264 (MPEG-4 AVC) as a representative proxy for contemporary lossy image/video compression techniques that are in common use within network-connected image/video devices and infrastructure, we examine the impact on performance across five discrete tasks: human pose estimation, semantic segmentation, object detection, action recognition, and monocular depth estimation. As such, within this study we include a variety of network architectures and domains spanning end-toend convolution, encoder-decoder, region-based CNN (R-CNN), dual-stream, and generative adversarial networks (GAN). Our results show a non-linear and non-uniform relationship between network performance and the level of lossy compression applied. Notably, performance decreases significantly below a JPEG quality (quantization) level of 15% and a H.264 Constant Rate Factor (CRF) of 40. However, retraining said architectures on precompressed imagery conversely recovers network performance by up to 78.4% in some cases. Furthermore, there is a correlation between architectures employing an encoder-decoder pipeline and those that demonstrate resilience to lossy image compression. The characteristics of the relationship between input compression to output task performance can be used to inform design decisions within future image/video devices and infrastructure.

ICPR 2020

On the Impact of Lossy Image and Video Compression on the Performance of Deep Convolutional Neural Network Architectures

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

In this talk we present a novel defogging technique, named CurL-Defog, aimed at minimizing the creation of unwanted artifacts during the defogging process. The majority of learning based defogging approaches rely on paired data (i.e., the same images with and without fog), where fog is artificially added to clear images: this often provides good results on mildly fogged images but does not generalize well to real difficult cases. On the other hand, the models trained with real unpaired data (e.g. CycleGAN) can provide visually impressive results but they often produce unwanted artifacts. In this paper we propose a curriculum learning strategy coupled with an enhanced CycleGAN model in order to reduce the number of produced artifacts, while maintaining state-of-the-art performance in terms of contrast enhancement and image reconstruction. We also introduce a new metric, called HArD (Hazy Artifact Detector) to numerically quantify the amount of artifacts in the defogged images, thus avoiding the tedious and subjective manual inspection of the results. The proposed approach compares favorably with state-of-the-art techniques on both real and synthetic datasets.

Towards Artifact-Free Image Defogging

In this work, we present a deep learning based framework to explore to replicate similar or even higher performance as costly LIDARs with our designed self-supervised network and a low-cost single-beam LIDAR.

Extending Single Beam Lidar To Full Resolution By Fusing with
Single Image Depth Estimation

We present a new measure of dependence suitable for time series forecasting: Partial Monotone Correlation (PMC) that generalizes Monotone Correlation. Unlike the Monotone Correlation, the new measure of dependence uses piecewise strictly monotone transformations that increase the value of the correlation coefficient. We explore its properties, its relationship with Monotone and Maximal Correlation, and present an algorithm that calculates it based on the Simultaneous Perturbation Stochastic Approximation method. We also demonstrate how to apply Partial Monotone Correlation for time series analysis and forecasting introducing Partial Monotone Autoregressive model (PMAR) of order 1. Its performance is then evaluated against the baseline of linear and nonlinear autoregressive models (AR, LSTAR) on 150 time series produced from 3 datasets: Yellow Taxi pickups, Citi Bike pickups, and Cellular Network hits. Overall, PMAR model outperforms the baseline with the average sMAPE of about 1.7-4% lower.

Partial Monotone Dependence

In this paper, we address the problem of utilizing web images in training robust joint embedding models for the image-text retrieval task. Prior webly supervised approaches directly leverage weakly annotated web images in the joint embedding learning framework. The objective of these approaches would suffer significantly when the ratio of noisy and missing tags associated with the web images is very high. In this regard, we propose a CP decomposition based tensor completion framework to refine the tags of web images by modeling observed ternary inter-relations between the sets of labeled images, tags, and web images as a tensor. To effectively deal with the high ratio of missing entries likely in our case, we incorporate intra-modal correlation as side information in the proposed framework. Our tag refinement approach combined with existing web supervised image-text embedding approaches provide a more principled way for learning the joint embedding models in the presence of significant noise from web data and limited clean labeled data. Experiments on benchmark datasets demonstrate that the proposed approach helps to achieve a significant performance gain in image-text retrieval.

Webly Supervised Image-Text Embedding with Noisy Tag Refinement

This paper targets to explore the inter-subject variations eliminated facial expression representation in the compressed video domain. Most of the previous methods process the RGB images of a sequence, while the off-the-shelf and valuable expression-related muscle movement already embedded in the compression format. In the up to two orders of magnitude compressed domain, we can explicitly infer the expression from the residual frames and possible to extract identity factors from the I frame with a pre-trained face recognition network. By enforcing the marginal independent of them, the expression feature is expected to be purer for the expression and be robust to identity shifts. We do not need the identity label or multiple expression samples from the same person for identity elimination. Moreover, when the apex frame is annotated in the dataset, the complementary constraint can be further added to regularize the feature-level game. In testing, only the compressed residual frames are required to achieve expression prediction. Our solution can achieve comparable or better performance than the recent decoded image based methods on the typical FER benchmarks with about 3$\times$ faster inference with compressed data.

Identity-aware Facial Expression Recognition in Compressed Video

Images acquired with standard digital cameras have
Bayer patterns and suffer from lens blur. A demosaicking step
is implemented in every digital camera, yet blur often remains
unattended due to computational cost and instability of deblurring
algorithms. Linear methods, which are computationally less
demanding, produce ringing artifacts in deblurred images. Complex
non-linear deblurring methods avoid artifacts, however, their
complexity implies offline application after camera demosaicking,
which leads to sub-optimal performance. In this work, we propose
a joint demosaicking deblurring and deringing network with a
light-weight architecture inspired by the alternating direction
method of multipliers. The proposed network has a transparent
and clear interpretation compared to other black-box data-driven
approaches. We experimentally validate its superiority over state-of-
the-art demosaicking methods with offline deblurring.

D3Net: Joint Demosaicking, Deblurring and Deringing

The single-image super-resolution (SISR) aims to recover a potential high-resolution image from its low-resolution version. Recently, deep learning-based methods have played a significant role in super-resolution field due to its effectiveness and efficiency. However, most of the SISR methods neglect the importance among the feature map channels. Moreover, they can not eliminate the redundant noises, making the output image be blurred. In this paper, we propose the residual subtraction and attention network (RSAN) for powerful feature expression and channels importance learning. More specifically, RSAN firstly implements one redundance removal module to learn noise information in the feature map and subtract noise through residual learning. Then it introduces the channel attention module to amplify high-frequency information and suppress the weight of effectless channels. Experimental results on extensive public benchmarks demonstrate our RSAN achieves significant improvement over the previous SISR methods in terms of both quantitative metrics and visual quality.

RSAN: Residual Subtraction and Attention Network for Single Image Super-Resolution

Image inpainting faces the challenging issue of the requirements on structure reasonableness and texture coherence. In this paper, we propose a two-stage inpainting framework to address this issue. The basic idea is to address the two requirements in two separate stages. Completed segmentation of the corrupted image is firstly predicted through segmentation reconstruction network, while fine-grained image details are restored in the second stage through an image generator. The two stages are connected in series as the image details are generated under the guidance of completed segmentation map that predicted in the first stage. Specifically, in the second stage, we propose a novel graph-based relation network to model the relationship existed in corrupted image. In relation network, both intra-relationship for pixels in the same semantic region and inter-relationship between different semantic parts are considered, improving the consistency and compatibility of image textures. Besides, contrastive loss is designed to facilitate the relation network training. Such a framework not only simplifies the inpainting problem directly, but also exploits the relationship in corrupted image explicitly. Extensive experiments on various public datasets quantitatively and qualitatively demonstrate the superiority of our approach compared with the state-of-the-art.

Image Inpainting with Contrastive Relation Network

The performance of an audio-visual sound source separation system is determined by its ability to separate audio sources given the images of the sources and the audio mixture. The goal of this study is to investigate the ability to learn the map-ping between the sounds and the images of instruments in the self-supervisied mix-and-seperate training paradigm used by state-of-the-art audio-visual sound source separation methods. Theoretical and empirical analyses illustrate that the self-supervised mix-and-separate training does not automatically learn the 1-to-1 correspondence between visual and audio signals, leading to low audio-visual object classification accuracy. Based on this analysis, a weakly-supervised method called Object-Prior is proposed and evaluated on two audio-visual datasets. The experimental results show that the Object-Prior method outperforms state-of-the-art baselines in the audio-visual sound source separation task. It is also more robust against asynchronized data, where the frame and the audio do not come from the same video, and recognizes musical instruments based on their sound with higher accuracy. This indicates that learning the 1-to-1 correspondence between visual and audio features of an instrument improves the effectiveness of audio-visual sound source separation.

Improving Mix-and-Separate Training in Audio-Visual Sound Source Separation with an Object Prior

Sound event recognition systems are rapidly becoming part of our life, since they can be profitably used in several vertical markets, ranging from audio security applications to scene classification and multi-modal analysis in social robotics. In the last years, a not negligible part of the scientific community started to apply Convolutional Neural Networks (CNNs) to image-based representations of the audio stream, due to their successful adoption in almost all the computer vision tasks.
In this paper, we carry out a detailed benchmark of various widely used CNN architectures and visual representations on a popular dataset, namely the MIVIA Audio Events database. Our analysis is aimed at understanding how these factors affect the sound event recognition performance with a particular focus on the false positive rate, very relevant in audio surveillance solutions. In fact, although most of the proposed solutions achieve a high recognition rate, the capability of distinguishing the events-of-interest from the background is often not yet sufficient for real systems, and prevent its usage in real applications. Our comprehensive experimental analysis investigates this aspect and allows to identify useful design guidelines for increasing the specificity of sound event recognition systems.

Premium content

On the Impact of Lossy Image and Video Compression on the Performance of Deep Convolutional Neural Network Architectures

Downloads

Next from ICPR 2020

Towards Artifact-Free Image Defogging

Similar lecture

HFP: Hardware-Aware Filter Pruning for Deep Convolutional Neural Networks Acceleration

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES