Video stabilization techniques are essential for most hand-held captured videos due to high-frequency shakes. Several 2D-, 2.5D-, and 3D-based stabilization techniques have been presented previously, but to the best of our knowledge, no solutions based on deep neural networks had been proposed to date. The main reason for this omission is shortage in training data as well as the challenge of modeling the problem using neural networks. In this paper, we present a video stabilization technique using a convolutional neural network. Previous works usually propose an off-line algorithm that smoothes a holistic camera path based on feature matching. Instead, we focus on low-latency, real-time camera path smoothing that does not explicitly represent the camera path and does not use future frames. Our neural network model, called StabNet, learns a set of mesh-grid transformations progressively for each input frame from the previous set of stabilized camera frames and creates stable corresponding latent camera paths implicitly. To train the network, we collect a dataset of synchronized steady and unsteady video pairs via a specially designed hand-held hardware. Experimental results show that our proposed online method performs comparatively to the traditional off-line video stabilization methods without using future frames while running about 10 times faster. More importantly, our proposed StabNet is able to handle low-quality videos, such as night-scene videos, watermarked videos, blurry videos, and noisy videos, where the existing methods fail in feature extraction or matching.
17 September 2020 in IEEE Transactions on Visualization and Computer Graphics
Video portraits are common in a variety of applications, such as videoconferencing, news broadcasting, and virtual education and training. We present a novel method to synthesize photorealistic video portraits for an input portrait video, automatically driven by a person's voice. The main challenge in this task is the hallucination of plausible, photorealistic facial expressions from input speech audio. To address this challenge, we employ a parametric 3D face model represented by geometry, facial expression, illumination, etc., and learn a mapping from audio features to model parameters. The input source audio is first represented as a high-dimensional feature, which is used to predict facial expression parameters of the 3D face model. We then replace the expression parameters computed from the original target video with the predicted one, and rerender the reenacted face. Finally, we generate a photorealistic video portrait from the reenacted synthetic face sequence via a neural face renderer. One appealing feature of our approach is the generalization capability for various input speech audio, including synthetic speech audio from text-to-speech software. Extensive experimental results show that our approach outperforms previous general-purpose audio-driven video portrait methods. This includes a user study demonstrating that our results are rated as more realistic than previous methods.
We introduce ShadowGAN, a generative adversarial network (GAN) for synthesizing shadows for virtual objects inserted in images. Given a target image containing several existing objects with shadows, and an input source object with a specified insertion position, the network generates a realistic shadow for the source object. The shadow is synthesized by a generator; using the proposed local adversarial and global adversarial discriminators, the synthetic shadow’s appearance is locally realistic in shape, and globally consistent with other objects’ shadows in terms of shadow direction and area. To overcome the lack of training data, we produced training samples based on public 3D models and rendering technology. Experimental results from a user study show that the synthetic shadowed results look natural and authentic.
Virtual reality (VR) offers an artificial, computer generated simulation of a real life environment. It originated in the 1960s and has evolved to provide increasing immersion, interactivity, imagination, and intelligence. Because deep learning systems are able to represent and compose information at various levels in a deep hierarchical fashion, they can build very powerful models which leverage large quantities of visual media data. Intelligence of VR methods and applications has been significantly boosted by the recent developments in deep learning techniques. VR content creation and exploration relates to image and video analysis, synthesis and editing, so deep learning methods such as fully convolutional networks and general adversarial networks are widely employed, designed specifically to handle panoramic images and video and virtual 3D scenes. This article surveys recent research that uses such deep learning methods for VR content creation and exploration. It considers the problems involved, and discusses possible future directions in this active and emerging research area.
31 May 2022 in Journal of Computer Science and Technology
Open-vocabulary querying in 3D space is challenging but essential for scene understanding tasks such as object localization and segmentation. Language embedded scene representations have made progress by incorporating language features into 3D spaces. However, their efficacy heavily depends on neural networks that are resource-intensive in training and rendering. Although recent 3D Gaussians offer efficient and high-quality novel view synthesis, directly embedding language features in them leads to prohibitive memory usage and decreased performance. In this work, we introduce Language Embedded 3D Gaussians, a novel scene representation for open-vocabulary query tasks. Instead of embedding high-dimensional raw semantic features on 3D Gaussians, we propose a dedicated quantization scheme that drastically alleviates the memory requirement, and a novel embedding procedure that achieves smoother yet high accuracy query, countering the multi-view feature inconsistencies and the high-frequency inductive bias in point-based representations. Our comprehensive experiments show that our representation achieves the best visual quality and language querying accuracy across current language embedded representations, while maintaining real-time rendering frame rates on a single desktop GPU.