Abstract:Video quality assessment (VQA) is an important processing task, aiming at predicting the quality of videos in a manner highly consistent with human judgments of perceived quality. Traditional VQA models based on natural image and/or video statistics, which are inspired both by models of projected images of the real world and by dual models of the human visual system, deliver only limited prediction performances on real-world user-generated content (UGC), as exemplified in recent large-scale VQA databases containing large numbers of diverse video contents crawled from the web. Fortunately, recent advances in deep neural networks and Large Multimodality Models (LMMs) have enabled significant progress in solving this problem, yielding better results than prior handcrafted models. Numerous deep learning-based VQA models have been developed, with progress in this direction driven by the creation of content-diverse, large-scale human-labeled databases that supply ground truth psychometric video quality data. Here, we present a comprehensive survey of recent progress in the development of VQA algorithms and the benchmarking studies and databases that make them possible. We also analyze open research directions on study design and VQA algorithm architectures.
Abstract:Advances in Single-vehicle intelligence of automated driving have encountered significant challenges because of limited capabilities in perception and interaction with complex traffic environments. Cooperative Driving Automation~(CDA) has been considered a pivotal solution to next-generation automated driving and intelligent transportation. Though CDA has attracted much attention from both academia and industry, exploration of its potential is still in its infancy. In industry, companies tend to build their in-house data collection pipeline and research tools to tailor their needs and protect intellectual properties. Reinventing the wheels, however, wastes resources and limits the generalizability of the developed approaches since no standardized benchmarks exist. On the other hand, in academia, due to the absence of real-world traffic data and computation resources, researchers often investigate CDA topics in simplified and mostly simulated environments, restricting the possibility of scaling the research outputs to real-world scenarios. Therefore, there is an urgent need to establish an open-source ecosystem~(OSE) to address the demands of different communities for CDA research, particularly in the early exploratory research stages, and provide the bridge to ensure an integrated development and testing pipeline that diverse communities can share. In this paper, we introduce the OpenCDA research ecosystem, a unified OSE integrated with a model zoo, a suite of driving simulators at various resolutions, large-scale real-world and simulated datasets, complete development toolkits for benchmark training/testing, and a scenario database/generator. We also demonstrate the effectiveness of OpenCDA OSE through example use cases, including cooperative 3D LiDAR detection, cooperative merge, cooperative camera-based map prediction, and adversarial scenario generation.
Abstract:Video and image quality assessment has long been projected as a regression problem, which requires predicting a continuous quality score given an input stimulus. However, recent efforts have shown that accurate quality score regression on real-world user-generated content (UGC) is a very challenging task. To make the problem more tractable, we propose two new methods - binary, and ordinal classification - as alternatives to evaluate and compare no-reference quality models at coarser levels. Moreover, the proposed new tasks convey more practical meaning on perceptually optimized UGC transcoding, or for preprocessing on media processing platforms. We conduct a comprehensive benchmark experiment of popular no-reference quality models on recent in-the-wild picture and video quality datasets, providing reliable baselines for both evaluation methods to support further studies. We hope this work promotes coarse-grained perceptual modeling and its applications to efficient UGC processing.
Abstract:Many objective video quality assessment (VQA) algorithms include a key step of temporal pooling of frame-level quality scores. However, less attention has been paid to studying the relative efficiencies of different pooling methods on no-reference (blind) VQA. Here we conduct a large-scale comparative evaluation to assess the capabilities and limitations of multiple temporal pooling strategies on blind VQA of user-generated videos. The study yields insights and general guidance regarding the application and selection of temporal pooling models. In addition, we also propose an ensemble pooling model built on top of high-performing temporal pooling models. Our experimental results demonstrate the relative efficacies of the evaluated temporal pooling models, using several popular VQA algorithms, and evaluated on two recent large-scale natural video quality databases. In addition to the new ensemble model, we provide a general recipe for applying temporal pooling of frame-based quality predictions.