||Special Issue on Graph Signal Processing: Foundations and Emerging Directions
- · Theoretical foundations for GSP: Advanced models for graphs, graph signals and graph filters
- · Nonlinear GSP
- · Beyond graph models: Hyper-based and tensor-based GSP
- · Statistical and robust GSP
- · Graph topology inference, including directed graphs and applications to causality
- · Machine learning for graph signals and geometric data
- · Applications of SP overdirected graphs to causal inference
- · Deep learning architectures for graph signals and geometric data
- · Algorithmic advances, distributed computations and large-scale graphs
- · Bioengineering, neuroscience and bioinformatics using GSP-tools
- · Communication, power, and transportation networks using GSP-tools
- · Finance, economics, and social networks using GSP-tools
- · Speech, image and video processing using GSP-tools
|Oct. 7th, 2019
||Data Driven Media Authentication and Forensics
- · Learning deep features relevant to low-level forensic analysis for problems like manipulation detection, identification of the social network of origin, camera model identification, or detection of artificially generated content.
- · Pro-active protection based on digital signatures, watermarking or other such integrity mechanisms based on machine learning.
- · Adoption of high-level vision to automate manual analysis that exposes physical inconsistencies, such as reflections, or shadows.
- · Media-phylogeny.
- · Addressing counter-forensic andadversarial attacks.
- · Forensics in the presence of in-camera processing such as HDR, video stabilization, neural imaging pipelines and advanced image fusion techniques.
- · Reconstruction of media genealogy.
- · Analysis and detection of imagery and videos created by new synthesis methods such as Generative models (GANs and VAEs).
- · Registration of media and their signatures in a central repository such as blockchains.
- · Accountability of forensics techniques.
- · Multimedia authorship attribution.
- · Accountable Machine-Learning techniques for Forensics.
- · Fairness, Accountability and Transparency in ML-based Forensics Methods.
|Nov. 1st, 2019
||Computer Vision for All Seasons: Adverse Weatherand Lighting Conditions
- · Image de-hazing (de-fogging), image de-raining and image de-snowing
- · Shadow removal, glare removal and reflection removal
- · Low-light image enhancement and HDR imaging
- · Style transfer and image translation across weather conditions, time of day, and seasons
- · Optical flow, depth estimation(from stereo or monocular),visual odometry, etc. in bad weather and at nighttime
- · Semantic scene understanding in bad weather and at nighttime
- · Domain adaptation from good weather/illumination conditions to adverse conditions
- · Learning with synthetic data for adverse weather/illumination conditions
- · Vision algorithms invariant to illumination, time of day, weather, and seasons
- · Datasets for bad weather and adverse lighting conditions
- · Fusing RGB cameras with other types of sensors to handle adverse conditions
- · New sensors and novel hardware setups for adverse weather and lighting conditions
- · Identification of visibility conditions
- · Robust vision algorithms against other adverse conditions
|Dec. 10th, 2019