BREAKING NEWS
LATEST POSTS
-
AI and the Law – Disney, Warner Bros. Discovery and NBCUniversal sue Chinese AI firm MiniMax
On Tuesday, the three media companies filed a lawsuit against MiniMax, a Chinese AI company that is reportedly valued at $4 billion, alleging “willful and brazen” copyright infringement
MiniMax operates Hailuo AI
-
Mariko Mori – Kamitate Stone at Sean Kelly Gallery
Mariko Mori, the internationally celebrated artist who blends technology, spirituality, and nature, debuts Kamitate Stone I this October at Sean Kelly Gallery in New York. The work continues her exploration of luminous form, energy, and transcendence.
-
Vimeo Enters into Definitive Agreement to Be Acquired by Bending Spoons for $1.38 Billion
The OG video hosting company is being acquired by Bending Spoons, the same company that bought WeTransfer and quietly changed terms to let them use your content for AI training.
-
ByteDance Seedream 4.0 – Super‑fast, 4K, multi image support
https://seed.bytedance.com/en/seedream4_0
➤ Super‑fast, high‑resolution results : resolutions up to 4K, producing a 2K image in less than 1.8 seconds, all while maintining sharpness and realism.
➤ At 4K, cost as low as 0.03 $ per generation.
➤ Natural‑language editing – You can instruct the model to “remove the people in the background,” “add a helmet” or “replace this with that,” and it executes without needing complicated prompts.
➤ Multi‑image input and output – It can combine multiple images, transfer styles and produce storyboards or series with consistent characters and themes.
-
OpenAI Backs Critterz, an AI-Made Animated Feature Film
https://www.wsj.com/tech/ai/openai-backs-ai-made-animated-feature-film-389f70b0
Film, called ‘Critterz,’ aims to debut at Cannes Film Festival and will leverage startup’s AI tools and resources.
“Critterz,” about forest creatures who go on an adventure after their village is disrupted by a stranger, is the brainchild of Chad Nelson, a creative specialist at OpenAI. Nelson started sketching out the characters three years ago while trying to make a short film with what was then OpenAI’s new DALL-E image-generation tool.
FEATURED POSTS
-
Vimeo Enters into Definitive Agreement to Be Acquired by Bending Spoons for $1.38 Billion
The OG video hosting company is being acquired by Bending Spoons, the same company that bought WeTransfer and quietly changed terms to let them use your content for AI training.
-
Photography Basics : Spectral Sensitivity Estimation Without a Camera
https://color-lab-eilat.github.io/Spectral-sensitivity-estimation-web/
A number of problems in computer vision and related fields would be mitigated if camera spectral sensitivities were known. As consumer cameras are not designed for high-precision visual tasks, manufacturers do not disclose spectral sensitivities. Their estimation requires a costly optical setup, which triggered researchers to come up with numerous indirect methods that aim to lower cost and complexity by using color targets. However, the use of color targets gives rise to new complications that make the estimation more difficult, and consequently, there currently exists no simple, low-cost, robust go-to method for spectral sensitivity estimation that non-specialized research labs can adopt. Furthermore, even if not limited by hardware or cost, researchers frequently work with imagery from multiple cameras that they do not have in their possession.
To provide a practical solution to this problem, we propose a framework for spectral sensitivity estimation that not only does not require any hardware (including a color target), but also does not require physical access to the camera itself. Similar to other work, we formulate an optimization problem that minimizes a two-term objective function: a camera-specific term from a system of equations, and a universal term that bounds the solution space.
Different than other work, we utilize publicly available high-quality calibration data to construct both terms. We use the colorimetric mapping matrices provided by the Adobe DNG Converter to formulate the camera-specific system of equations, and constrain the solutions using an autoencoder trained on a database of ground-truth curves. On average, we achieve reconstruction errors as low as those that can arise due to manufacturing imperfections between two copies of the same camera. We provide predicted sensitivities for more than 1,000 cameras that the Adobe DNG Converter currently supports, and discuss which tasks can become trivial when camera responses are available.
