Why My AI Videos look Ultra Realistic - Higgsfield AI

Importance of Cinematic Images 00:26

"Most of your effort should be in generating your image because that will be your reference and the ingredient for your cinematic video."

• To create truly cinematic videos, a significant focus should be placed on generating high-quality images, as they serve as the foundational reference for animations.

• The speaker emphasizes that a considerable amount of time—up to 99%—should be dedicated to perfecting images through concepting and storyboarding before generating video content.

• It's essential to utilize the right tools for creating cinematic images, including Nano Banana Pro, Nano Banana 2, and Higgs Field Soul.

Key Elements of Cinematic Images 01:28

"Lighting is an insanely important aspect. On film sets, you have entire lighting crews trying to light that perfect shot."

• Effective cinematic images hinge on essential elements such as lighting, composition, depth, and color.

• Proper lighting adds motivation, direction, shadow, contrast, highlight, and depth, transforming flat images into captivating visuals.

• Composition is crucial, with the rule of thirds serving as a guide for placing subjects. Properly positioning important focuses can significantly enhance visual storytelling.

• Adding foreground, background, and mid-ground details creates the necessary depth, while color adjustments can greatly affect the final cinematic feel.

Using Tools for Image Generation 03:32

"So in this video, we will be using Higgsfield Soul and Nano Banana 2 to make our images look way more cinematic."

• The video elaborates on utilizing tools like Higgsfield Soul and Nano Banana 2 to achieve highly cinematic images.

• Soul 2 provides the option to incorporate various styles and maintain character consistency, further enhancing the uniqueness of the generated images.

• Capturing your character for the AI involves training it with multiple images that showcase different angles and expressions for better accuracy.

Creating Unique Characters 05:48

"In order to train your Soul ID, it's recommended that we have over 20 different images."

• Developing unique characters for AI generation is crucial for achieving personalized cinematic shots.

• By utilizing a set of various images, you can effectively train the AI model to depict a consistent character representation.

• The speaker offers a downloadable document with prompts that aid in generating a diverse set of images necessary for character training.

The Workflow for Image Creation 07:00

"The first step is using Soul 2 for image referencing."

• Establishing a mood board is the initial step, allowing inspiration to guide the vision for the project.

• Reference images are imported into Soul 2, where the character model is utilized to generate images based on the chosen references.

• The workflow also includes refining generated images by providing prompts for specific details, ensuring that the AI maintains character identity while making adjustments to outfit or setting.

Generating Images with Higgs Field Soul 08:52

"If you're generating anything using Higgs Field Soul, it leaves behind the prompt that it is using."

• The video discusses how to use Higgs Field Soul to generate images, highlighting the ability to capture the prompts that create specific results.

• The process begins by selecting a reference image from Pinterest, which the creator uses as a basis for their own generation.

• After inputting the character, Veronica, the creator generates an initial image, but finds it insufficient, prompting a return to the drawing board.

• By deleting the reference image and pasting in the copied prompt, the creator experiments with different styles, ultimately using a digital camera setting to refine the image.

Detailing and Iterating Image Generation 09:56

"That is how you keep reiterating your images; you keep adding stuff or removing stuff, making it better with each generation."

• The creator emphasizes the importance of incorporating additional details to enhance the image quality further.

• They mention using descriptive prompts such as having a woman in a ruined wedding dress, crying with mascara running down her face, to evoke emotion and visual depth.

• The iterative process is essential for developing cinematic images that resonate with viewers.

Using References and Prompt Adjustments 10:40

"Essentially, what I did is I changed up the prompt to match my style."

• The video explains how the creator uses various reference images and modifies prompts to personalize the generated results.

• Emphasizing attention to detail, they adjust specifics like the type of mug to reflect their preferences, showcasing a method for aligning generated images with personal branding.

• The creator illustrates this through comparisons of initial and refined images, demonstrating how the prompts lead to desired results.

The Role of Emotion in Cinematic Videos 11:49

"Adding emotions in our scene is what makes it relatable to the viewer."

• The creator highlights the significance of emotion in creating engaging and cinematic scenes, urging future creators to think about the feelings they want to evoke in viewers.

• They elaborate on how different colors impact emotion, with red representing intensity and blue signaling calmness, affecting how scenes are perceived.

• Camera directions, including zooming in or out, further contribute to the emotional weight of a scene, influencing viewer involvement and engagement.

Editing for Cinematic Impact 14:40

"Now, what you want to do is just generate it again."

• In the editing phase, the creator focuses on assembling clips, making quick cuts to eliminate unnecessary footage or errors.

• They advise adding overlays, such as a luma key or Gaussian blur effect, for a more cinematic appearance, noting this technique often enhances video quality and viewer experience.

• Throughout the process, the creator reflects on the cumulative effort required to achieve polished, cinematic videos, showcasing the reward of dedication and creativity in filmmaking.