BREAKING NEWS
LATEST POSTS
-
HumanDiT – Pose-Guided Diffusion Transformer for Long-form Human Motion Video Generation
https://agnjason.github.io/HumanDiT-page
By inputting a single character image and template pose video, our method can generate vocal avatar videos featuring not only pose-accurate rendering but also realistic body shapes.
-
DynVFX – Augmenting Real Videoswith Dynamic Content
Given an input video and a simple user-provided text instruction describing the desired content, our method synthesizes dynamic objects or complex scene effects that naturally interact with the existing scene over time. The position, appearance, and motion of the new content are seamlessly integrated into the original footage while accounting for camera motion, occlusions, and interactions with other dynamic objects in the scene, resulting in a cohesive and realistic output video.
https://dynvfx.github.io/sm/index.html
-
ByteDance OmniHuman-1
https://omnihuman-lab.github.io
They propose an end-to-end multimodality-conditioned human video generation framework named OmniHuman, which can generate human videos based on a single human image and motion signals (e.g., audio only, video only, or a combination of audio and video). In OmniHuman, we introduce a multimodality motion conditioning mixed training strategy, allowing the model to benefit from data scaling up of mixed conditioning. This overcomes the issue that previous end-to-end approaches faced due to the scarcity of high-quality data. OmniHuman significantly outperforms existing methods, generating extremely realistic human videos based on weak signal inputs, especially audio. It supports image inputs of any aspect ratio, whether they are portraits, half-body, or full-body images, delivering more lifelike and high-quality results across various scenarios.
-
Conda – an open source management system for installing multiple versions of software packages and their dependencies into a virtual environment
https://anaconda.org/anaconda/conda
https://docs.conda.io/projects/conda/en/latest/user-guide/getting-started.html
NOTE The company recently changed their TOS and this service now incurs into costs for teams above a threshold.
Use MicroMamba instead.
-
Vashi Nedomansky – Shooting ratios of feature films
In the Golden Age of Hollywood (1930-1959), a 10:1 shooting ratio was the norm—a 90-minute film meant about 15 hours of footage. Directors like Alfred Hitchcock famously kept it tight with a 3:1 ratio, giving studios little wiggle room in the edit.
Fast forward to today: the digital era has sent shooting ratios skyrocketing. Affordable cameras roll endlessly, capturing multiple takes, resets, and everything in between. Gone are the disciplined “Action to Cut” days of film.https://en.wikipedia.org/wiki/Shooting_ratio
-
General OCR Theory – Towards OCR-2.0 via a Unified End-to-end Model – HF Transformers implementation
https://huggingface.co/stepfun-ai/GOT-OCR-2.0-hf
GOT-OCR2 works on a wide range of tasks, including plain document OCR, scene text OCR, formatted document OCR, and even OCR for tables, charts, mathematical formulas, geometric shapes, molecular formulas and sheet music.
-
QNTM – Developer Philosophy
- Avoid, at all costs, arriving at a scenario where the ground-up rewrite starts to look attractive
- Aim to be 90% done in 50% of the available time
- Automate good practice
- Think about pathological data
- There is usually a simpler way to write it
- Write code to be testable
- It is insufficient for code to be provably correct; it should be obviously, visibly, trivially correct
-
Arminas Valunas – “Coca-Cola: Wherever you are.”
Arminas created this using Juggernaut Xl model and QR Code Monster SDXL ControlNet.
His pipeline:
Static Images – Forge UI.
Upscaled with Leonardo AI universal upscaler.
Animated with Runway ML and Minimax.
Video upscale – Topaz Video AI.
Composited in Adobe Premiere.
Juggernaut Xl download here:
https://civitai.com/models/133005/juggernaut-xl
QR Code Monster SDXL:
https://civitai.com/models/197247?modelVersionId=221829
FEATURED POSTS
-
Pika.art – an AI for creating videos from stills
“It converts simple text instructions into captivating videos, in seconds.
The story behind this AI is fascinating: A team of four engineers, led by Demi Guo and Chenlin Meng, was born with a clear vision: to transform video creation.
After raising $55 million, Pika Labs initially focused on Japanese anime-style animations before expanding into 3D animation”
-
What Is The Resolution and view coverage Of The human Eye. And what distance is TV at best?
https://www.discovery.com/science/mexapixels-in-human-eye
About 576 megapixels for the entire field of view.
Consider a view in front of you that is 90 degrees by 90 degrees, like looking through an open window at a scene. The number of pixels would be:
90 degrees * 60 arc-minutes/degree * 1/0.3 * 90 * 60 * 1/0.3 = 324,000,000 pixels (324 megapixels).At any one moment, you actually do not perceive that many pixels, but your eye moves around the scene to see all the detail you want. But the human eye really sees a larger field of view, close to 180 degrees. Let’s be conservative and use 120 degrees for the field of view. Then we would see:
120 * 120 * 60 * 60 / (0.3 * 0.3) = 576 megapixels.
Or.
7 megapixels for the 2 degree focus arc… + 1 megapixel for the rest.
https://clarkvision.com/articles/eye-resolution.html
Details in the post
-
About color: What is a LUT
http://www.lightillusion.com/luts.html
https://www.shutterstock.com/blog/how-use-luts-color-grading
A LUT (Lookup Table) is essentially the modifier between two images, the original image and the displayed image, based on a mathematical formula. Basically conversion matrices of different complexities. There are different types of LUTS – viewing, transform, calibration, 1D and 3D.
