Resources

Here is a non-exhaustive list of resources used by the group, made available to the research community.

Datasets

• Large scale dataset of HEVC-encoded video sequences

• Precomputed VQMs for the ITS4S dataset (accompanying the paper “Full Reference Video Quality Measures Improvement using Neural Networks” (alt link) published in IEEE ICASSP 2020)

• Dataset/AI Observers developed with the human-in-the-loop approach (accompanying the paper “Training the DNN of a Single Observer by Conducting Individualized Subjective Experiments” (alt link) published in QoMEX 2023)

Software

List of software tools freely available for research purposes, developed in the context of the groups’ activities or made available for the group:

• Simulation software for the paper “On the Link Between Subjective Score Prediction and Disagreement of Video Quality Metrics” (published in IEEE Access, 2021)

• Reproducible research framework (accompanying the journal paper “Reproducible research framework for objective video quality measures using a large-scale database approach” published (open access) in SoftwareX in 2018)

• Robust AVC decoder

• PVQM quality measure

• AI Observers models (trained Matlab NNs) presented in the paper Predicting individual quality ratings of compressed images through deep CNNs-based artificial observers (published in Signal Processing: Image Communication, 2023) (Download Preprint)

Also, take a look to the VQEG Tools and Subjective Labs Setup website that contains other resources and links.