最新消息最新消息

近日,有消息称字节跳动推出了一款名为Boximator的视频生成模型,据称其可以通过文本精准控制视频中的人物或物体的动作。字节跳动相关人士对此作出回应,称Boximator仍处于技术探索阶段,距离成熟仍有差距,无法作为产品落地。

相关人士表示,Boximator在画面质量、保真率和生成视频长度等方面,与国外领先的生成模型仍存在明显差距。他们将继续努力,提升生成视频的各项指标,争取尽快实现产业化。

Title: ByteDance denies Boximator is mature
Keywords: Boximator, video generation, immature
News content: Recently, there was news that ByteDance launched a video generation model called Boximator, which can supposedly precisely control the actions of people or objects in videos through text. ByteDance insiders responded that Boximator is still at the technology exploration stage and there is still a gap before maturity, unable to land as a product.

The insiders said that in terms of image quality, fidelity rate and length of generated videos, Boximator still has significant gaps compared with leading foreign generation models. They will continue to make efforts to improve various indicators of generated videos and strive to achieve industrialization as soon as possible.

【来源】https://finance.sina.com.cn/tech/shenji/2024-02-20/doc-inairynk7158099.shtml

Views: 1

发表回复

您的邮箱地址不会被公开。 必填项已用 * 标注