TikTok's parent company launches technology to challenge Google.

ByteDance 's new tool confidently surpasses all competitors. Photo: ByteDance .

ByteDance, the parent company of TikTok and Douyin, has officially introduced Seedance 1.0, an artificial intelligence (AI) tool capable of creating videos from text and images. This is seen as a significant step for the Chinese technology giant in the race to develop AI-powered content creation platforms.

The company recently published a detailed study on Seedance 1.0. According to the study, this tool is designed to convert simple instructions into high-quality video without requiring detailed scripts or complex commands. It handles not only individual shots but also combines multiple camera angles, ensures smooth transitions, and maintains character consistency throughout the video.

“We’ve found a way to separate spatial and temporal information in video. This technology uses a special method to ‘encode’ location, helping AI learn to create videos from both text and images within the same model. As a result, AI can automatically create videos with multiple different shots smoothly,” ByteDance stated in the research paper.

ByteDance confidently claims that Seedance 1.0 surpasses existing AI video creation tools on the market, particularly in its ability to closely follow user ideas, image sharpness, and natural character movement.

According to Artificial Analysis , a platform specializing in analyzing and evaluating the performance of AI models, Seedance 1.0 has surpassed other AI video creation tools such as Google's Veo 3, Kuaishou's Kling 2.0, and OpenAI's Sora. This tool demonstrated superior performance in both text-to-video and image-to-video conversion tasks.

The company also revealed that Seedance 1.0 was trained on a massive database of videos, collected from publicly available and licensed sources. The training videos underwent a rigorous filtering process to remove violent or sensitive content.

Many believe that the data primarily comes from TikTok and Douyin, two platforms operated by ByteDance itself.

The Seedance 1.0 training process is divided into several stages: initially learning from rich image and video data, then continuing to learn in-depth about transition techniques in various styles.

Humans also play a crucial role in the training process, as engineers select high-quality videos for the model to learn from. The training loop continues until Seedance 1.0 can independently select the optimal result from among many videos generated on demand.

Currently, Seedance 1.0 limits video length to a maximum of 5 seconds (compared to 8 seconds for Veo 3). However, its standout advantage is its fast processing speed: it only takes 41 seconds to create a Full HD video. One drawback of Seedance 1.0 is that it doesn't yet support automatic sound overlay like its Google counterpart.

ByteDance plans to release this tool soon for both general users and professional content creators, serving the needs of producing promotional videos or short content for social media.

Prior to Seedance 1.0, ByteDance had developed AI video creation tools such as OmniHuman, Goku, and Jimeng AI. However, Seedance 1.0 is the first product that the company confidently claims can surpass its competitors in AI-powered video creation capabilities.

Source: https://znews.vn/cong-ty-me-tiktok-ra-mat-cong-nghe-thach-thuc-google-post1562025.html