In this paper, we propose a novel text-based talking-head video generation framework that synthesizes high-fidelity facial expressions and head motions in accordance with contextual sentiments as well as speech rhythm and pauses. To be specific, our
framework consists of a speaker-independent stage and a speaker-specific stage. In the speaker-independent stage, we design three parallel networks to generate animation parameters of the mouth, upper face, and head from texts, separately. In the speaker-specific stage, we present a 3D face model guided attention network to synthesize videos tailored for different individuals. It takes the animation parameters as input and exploits an attention mask to manipulate facial expression changes for the input individuals. Furthermore, to better establish authentic correspondences between visual motions (i.e., facial expression changes and head movements) and audios, we leverage a high-accuracy motion capture dataset instead of relying on long videos of specific individuals. After attaining the visual and audio correspondences, we can effectively train our network in an end-to-end fashion. Extensive experiments on qualitative and quantitative results demonstrate that our algorithm achieves high-quality photo-realistic talking-head videos including various facial expressions and head motions according to speech rhythms and outperforms the state-of-the-art.
A novel regime of self-compression is proposed for plasma-based backward Raman amplification(BRA) upon flying focus. By using a pumping focus moving with a speed equal to the group velocity of stimulated Raman backscattering(SRBS), only a short part
of SRBS which does always synchronize with the flying focus can be amplified. Due to the asymmetrical amplification, the pulse can be directly compressed in the linear stage of BRA. Therefore, instead of a short pulse, the Raman spontaneous or a long pulse can seed the BRA amplifiers. The regime is supported by the 2D particle-in-cell(PIC) simulation without a seed, presenting that the pump pulse is compressed from 26ps to 116fs, with an output amplitude comparable with the case of a well-synchronized short seed. This method provides a significant way to simplify the Raman amplifiers and overcome the issue of synchronization jitter between the pump and the seed.
Although many methods perform well in single camera tracking, multi-camera tracking remains a challenging problem with less attention. DukeMTMC is a large-scale, well-annotated multi-camera tracking benchmark which makes great progress in this field.
This report is dedicated to briefly introduce our method on DukeMTMC and show that simple hierarchical clustering with well-trained person re-identification features can get good results on this dataset.
