What You May Have Missed this week:

1. Img to 3D Scene with LucidDreamer

2. A new LipSync / Avatar Talking tool

3. Consistent style without fine-tuning!

4. Control the generation

5. THE MOST MINDBLOWING DEMO I SEE THE LAST 2 WEEKS!

6. Our latest tutorial about Pika, the revolution for AI Films!

7. Our latest video. Grok Inside!

8. The pitch deck of the week: Mistral for their seed ($111M)

9. Ai Film Festival

1. Img to 3D Scene with LucidDreamer

Now, with LucidDreamer, you can generate a 3D scene with just one 2D image and a text prompt.
Then you can zoom and navigate in the scene.
It’s just incredible! The tech is new, and it’ll better much better in 2024 but it’s already more than usable!

You can play with it on HuggingFace.

2. A new LipSync / Avatar Talking tool

All existing solutions for LipSync are not good enough to make AI Films, but it could change with VividTalk. The results are better than the other solutions and seem working well with non-realistic faces!

3. Consistent style without fine-tuning!

Google just released StyleAdapter, a wonderful paper AND code! Consistent style generation withOUT fine-tuning. It could be a good challenger for IPAdapter solution!

4. Control the generation

"Readout Guidance" is an innovative approach designed to enhance the capabilities of text-to-image diffusion models. It is a significant advancement in the field of AI-generated imagery, providing more control, efficiency, and versatility in creating images that align closely with user-defined criteria and textual descriptions.

It picks up specific properties from the images being generated. This includes basic properties like the pose of a figure, the depth perception in the image, or the outline of objects. They can also understand more complex relations between multiple images, like how similar they are in appearance or how elements correspond between them.

5. THE MOST MINDBLOWING DEMO I SEE THE LAST 2 WEEKS!

With Animate Anyone, we’ll be able to animate any character with any movement! Create your character with Stable Diffusion or MidJourney, record or find a video of a dance, running, or anything else… and you have your character dancing, running…

6. Our latest tutorial:

The French version is available on our YT channel: https://www.youtube.com/@Fictions-ai

7. Fun moment: the “What if” video (Parody of iconic characters):

8. Pitch deck of Mistral for their seed ($111M):

8. AI Film Festivals

Two weeks ago, we organized the first AI Film Festival in France.
We were sold out with more than 200 people in the theater.
More than 100 videos in competition and a great jury (more information on https://www.aiff-montpellier.com/). You can watch the replay here:

And if you like AI Movie, Festival & Mashup, a good initiative exists: Screenshot shup:
https://comet-courts.fr/mashup-film-festival-cinemia/
With 35 minutes of projection, available online (and zero overlap with the AIFF Montpellier), you can have a large overview of what exists in AI Movies!

AI moves fast – don't get left behind! Let us be your guide.

Follow us on social networks: