Image and video synthesis are closely related areas aiming at generating content from noise. We present a framework that leverages contemporary image generators to render high-resolution videos.

MOTSynth is a huge dataset for pedestrian detection and tracking in urban scenarios created by exploiting the highly photorealistic video game Grand Theft Auto V developed by Rockstar North.

Developing and testing algorithms for autonomous vehicles in real world is an expensive and time consuming process. Also, in order to utilize recent advances in machine intelligence and deep learning we need to collect a large amount of annotated training data in a variety of conditions and environments. We present a new simulator built on Unreal Engine that offers physically and visually realistic simulations for both of these goals. Our simulator includes a physics engine that can operate at a high frequency for real-time hardware-in-the-loop (HITL) simulations with support for popular protocols (e.g. MavLink).

In this work, we demonstrate the large potential of synthetic data for analyzing and reducing the negative effects of dataset bias on deep face recognition systems.

10k images of 50 animation sequences of a diverse set of animated body poses, including motions in both the upper and lower body

CARLA, an open-source simulator for autonomous driving research. CARLA has been developed from the ground up to support development, training, and validation of autonomous urban driving systems. In addition to open-source code and protocols, CARLA provides open digital assets (urban layouts, buildings, vehicles) that were created for this purpose and can be used freely.

Crowd behavior analysis is a challenging task in computer vision, mainly due to the high complexity of the interactions between groups and individuals. We establish the first baseline of crowd characterization with an extensive evaluation on shallow and deep methods. This characterization is expected to be useful in multiple crowd analysis circumstances, we present a new deep architecture for crowd characterization and demonstrate its application in the context of anomaly classification.

10k images of a diverse set of humans with the camera focused on the head and shoulder, in multiple environments and camera angles

This paper introduces an updated version of the well-known Virtual KITTI dataset which consists of 5 sequence clones from the KITTI tracking benchmark. In addition, the dataset provides different variants of these sequences such as modified weather conditions (e.g. fog, rain) or modified camera configurations (e.g. rotated by 15 degrees).

Recently, counting the number of people for crowd scenes is a hot topic because of its widespread applications (e.g. video surveillance, public security). It is a difficult task in the wild: changeable environment, large-range number of people cause the current methods can not work well. In addition, due to the scarce data, many methods suffer from over-fitting to a different extent. To remedy the above two problems, firstly, we develop a data collector and labeler, which can generate the synthetic crowd scenes and simultaneously annotate them without any manpower. Secondly, we propose two schemes that exploit the synthetic data to boost the performance of crowd counting in the wild.

We present a method that achieves state-of-the-art results for synthesizing novel views of complex scenes by optimizing an underlying continuous volumetric scene function using a sparse set of input views. Our algorithm represents a scene using a fully connected (non-convolutional) deep network, whose input is a single continuous 5D coordinate (spatial location (x,y,z) and viewing direction (θ,ϕ)) and whose output is the volume density and view-dependent emitted radiance at that spatial location.

This paper presents a unified multimodal pre-trained model called N\"UWA that can generate new or manipulate existing visual data (i.e., images and videos) for various visual synthesis tasks.

Cameras are becoming ubiquitous in the Internet of Things (IoT) and can use face recognition technology to improve context. There is a large accuracy gap between today’s publicly available face recognition systems and the state-of-the-art private face recognition systems. This paper presents our OpenFace face recognition library that bridges this accuracy gap

We present a benchmark suite for visual perception. The benchmark is based on more than 250K high-resolution video frames, all annotated with ground-truth data for both low-level and high-level vision tasks, including optical flow, semantic instance segmentation, object detection and tracking, object-level 3D scene layout, and visual odometry. Ground-truth data for all tasks is available for every frame.

Deep learning for human action recognition in videos is making significant progress but is slowed down by its dependency on expensive manual labeling of large video collections. In this work, we investigate the generation of synthetic training data for action recognition

10k images of a diverse set of static body poses, including motions in fingers and hands, as well as both the upper and lower body

StyleGAN2 is a generative adversarial network that builds on StyleGAN with several improvements. First, adaptive instance normalization is redesigned and replaced with a normalization technique called weight demodulation. Secondly, an improved training scheme upon progressively growing is introduced, which achieves the same goal - training starts by focusing on low-resolution images and then progressively shifts focus to higher and higher resolutions - without changing the network topology during training.

SURREAL (Synthetic hUmans foR REAL tasks) is a large-scale person dataset that generates photorealistic synthetic images with labeling for human part segmentation and depth estimation, producing 6.5M frames in 67.5K short clips (about 100 frames each) of 2.6K action sequences with 145 different synthetic subjects. To ensure realism, the synthetic bodies are created using the SMPL body model, whose parameters are fit by the MoSh method given raw 3D MoCap marker data.

This dataset contains synthetic images which are obtained from an environment created in the Unity3D framework, called VEIS (Virtual Environment for Instance Segmentation). we used the Unity3D game engine, in which one can manually design scenes with common urban structures and add freely-available 3D objects, representing foreground classes, to the scene.

Virtual KITTI is a photo-realistic synthetic video dataset designed to learn and evaluate computer vision models for several video understanding tasks: object detection and multi-object tracking, scene-level and instance-level semantic segmentation, optical flow, and depth estimation.

Due to the advances in deep reinforcement learning and the demand of large training data, virtual-to-real learning has gained lots of attention from computer vision community recently. As state-of-the-art 3D engines can generate photo-realistic images suitable for training deep neural networks, researchers have been gradually applied 3D virtual environment to learn different tasks including autonomous driving, collision avoidance, and image segmentation, to name a few.

Fisheye cameras are commonly employed for obtaining a large field of view in surveillance, augmented reality and in particular automotive applications. In spite of their prevalence, there are few public datasets for detailed evaluation of computer vision algorithms on fisheye images. We release the first extensive fisheye automotive dataset, WoodScape, named after Robert Wood who invented the fisheye camera in 1906. WoodScape comprises of four surround view cameras and nine tasks including segmentation, depth estimation, 3D bounding box detection and soiling detection.