Context:
Recently, OpenAI, the creator of chatbot ChatGPT, has introduced a Generative Artificial Intelligence (GenAI) model named Sora OpenAI.
About Sora OpenAI:
Sora OpenAI in Japanese means sky, an imagery that evokes ‘limitless creative potential’.
- SORA OpenAI is a diffusion based AI model developed by OpenAI, specialising in transforming textual prompts into vivid and realistic video scenes.
-
- It leverages a diffusion model and transformer architecture to understand and simulate the physical world in motion.
- It is currently unavailable for general use as OpenAI is focusing on implementing safety protocols and gathering feedback from visual artists and filmmakers.
-
Features of Sora OpenAI
-
- Text to Video Capabilities: It can create videos lasting up to one minute, ensuring exceptional visual quality while following user instructions.
- Generating Complex Scenes: It crafts elaborate scenes featuring multiple characters, diverse motions, and precise details of both the subjects and backgrounds.
- Create Dynamic Impressions and EngAgeing Characters: Proficient in comprehending real-world object functionalities and accurately interpreting instructions.
- Multishot Avatar Production: It showcases the ability to generate multiple shots within a single video, maintaining consistency in characters and visual style.
| Other Companies apart from Open AI have ventured into the text-to-video space : Google’s Lumiere, Runway, pika etc. |
-
- Filters : To block prompt requests that mention violent, sexual or hateful language, as well as images of well-known personalities.
Also Read: What Is Deepfake Technology?
Limitations of Sora OpenAI
- Limited Dataset: If the dataset lacks certain types of scenes or visual variations, it may produce videos that lack realism or exhibit strange artefacts.
- Complex Scenes: Generating realistic videos becomes increasingly challenging when scenes involve complex interactions, intricate details, or dynamic elements.
- Temporal Consistency: It might encounter difficulties in ensuring smooth transitions and coherence between consecutive frames, leading to jarring or unnatural-looking sequences.
- Real-time Performance: Depending on the hardware and computational resources available, generating videos with Sora in real-time may pose challenges.
- Ethical Considerations: In accurately discerning between real and manipulated content could exacerbate these concerns if not carefully addressed.
- Domain Specificity: Its performance may vary across different domains or types of videos. It may excel in certain contexts while struggling in others.
Generative AI:
- Generative AI is a type of AI technology that can produce various types of content, including text, imagery, audio and synthetic data.
- It utilizes deep learning, neural networks, and machine learning techniques to enable computers to produce content that closely resembles human-created output autonomously.
- Example: ChatGPT, DALL-E, and Bard.
Diffusion Based AI model:
- These are a class of generative AI models that generate high-resolution images of varying quality.
- They work by gradually adding Gaussian noise to the original data in the forward diffusion process and then learning to remove the noise in the reverse diffusion process.
|
Also Read: Global Partnership On Artificial Intelligence – GPAI
News Source : The Hindu
February 19, 2024