Course Curriculum

AI movie making

6sections
30lectures
1h 56mtotal

What you’ll learn

  • Nanobanana, Sora, Capcut, OpenArt and the new age of AI media - 2026 updated, clear, and honest

  • Learn the AI video making tool box with updated information

  • Learn the flow to generate AI videos, mostly short videos

  • How to generate videos with custom characters and cameos

  • How to make longer scripts and long format videos with proper prompting habits

01

Introduction

6 lectures • 33min

  • Pro upgrade? which one first? Sora, Capcut, Gemini, Nanobana, and Openart?

    Compare Capcut, Sora, Gemini, Google View Flow, Nano Banana, and OpenArt to reveal the current ai video and movie making landscape, including design, ai video, editing, and storytelling capabilities.

    6:34

  • NanoBanana? is it free? what is the big deal new feature?

    I will discuss the most important features that people care about.  No it is not free, but you can ease in. Top features: 1. Image to video from one image 2. Swapping cloth on any picture (you can type in I want this character to wear this" and it will wear this. 3. Making great designs you can not make yourself, such as Youtube thumbnails.

    2:15

  • The pros and cons of Sora 2 by OpenAI

    Assess the pros and cons of Sora 2 by OpenAI, highlighting prompt engineering challenges, policy constraints, and the evolving landscape of AI tools for movie making.

    2:23

  • Best realistic avatar maker? a test of Sora and OpenArt

    Compare Sora two and Open Art for realistic custom characters, weighing selfie-based accuracy and sound against Open Art's image-based models and limits on longer stories.

    10:26

  • Tool box overview November 2025

    Explore the rapid state-of-the-art in AI movie making, from AI-generated cameos and avatars to mobile and PC tools like Sora and CapCut, guided by prompt engineering and practical trade-offs.

    9:22

  • These 3 features of nanobanana is life saver for Youtuber

    Explore three nano banana pro features for YouTubers: prompt-driven outfit swaps, image-to-video conversion, and automated YouTube thumbnail creation, with a note on 1–2 minute generation times.

    1:39
02

Individual softwares and basic use cases

8 lectures • 28min

  • Sora 2 - part 1

    Learn how to use Sora to create videos from prompts, review drafts, and post content, using the search box, leaderboard, and mobile access, while noting free access and potential costs.

    3:22

  • Making realistic avatar 10 sec video from Sora 2

    Learn how Sora two enables creating a realistic avatar in 10 seconds for short videos, and compare it with Nano Banana as a graphic designer and video magic maker.

    1:44

  • OpenArts - part 1

    Explore OpenArts, an open AI interface for stories, images, videos, and audio with character training. Learn about pricing tiers from essential to wonder and token-based access.

    4:14

  • OpemArt Unique Realistic Character and AI scripted story (800 credits 30 seconds

    Explore open art's tools to build characters, stories, videos, or audio with lip sync and training. Create ai-driven narratives and campus videos, while considering credits, costs, and limitations.

    7:38

  • All fake, all beautiful - a single image used for Capcut AI video

    Use a single image to craft a CapCut AI video that blends fake and beautiful visuals for AI movie making.

    1:10

  • NanoBanana Pro quick start November 2025

    Explore nano banana pro's quick start: transform a single image into a video, swap outfits with prompts, and generate realistic pictures, accessible via CapCut or Google, for paid subscribers.

    6:41

  • AI image to video engine tested - Seedance, VEO, Kling, Wan, Hailuo, Pixverse

    Directly compare several image-to-video algorithms using the same prompt and image, evaluating Chinese models alongside American models like Google and Sora for character consistency.

    2:28

  • Copyright free people face at thispersondoesnotexist.com

    Goto thispersondoesnotexist.com to look for high quality pictures of persons that do not exist on earth. If you want to avoid copyright issue and find handsome characters for your video or movie, here is a good place.  It is endless and all free.

    0:37
03

Various examples and recipes

5 lectures • 12min

  • Making vertical avatar speech videos on Capcut

    Learn to create vertical avatar speeches in CapCut using ai video maker, select Sir Arthur, type what he should say, and export as shorts for TikTok or YouTube Shorts.

    2:56

  • Using "AI Video" in Capcut to generate short video from a single picture (fee)

    Turn a single picture into a short video using CapCut's AI video feature; simply provide a picture and let the AI generate a short clip you can save.

    1:12

  • Capcut AI video from single picture with prompt direction experiment

    Explore CapCut's AI video tool to generate a video from a single image, using prompts to direct actions like looking left or toward the sky.

    3:13

  • Change model clothing in nano banana pro

    Learn to change a model's clothing to business casual with a black blazer using nano banana pro and capcut pro via the web interface, and see the resulting edit.

    1:59

  • Use your cloth on a glamour model (for marketing) in capcut

    Learn to overlay clothing onto an AI fashion model in CapCut for marketing, select a garment, apply it to a model, and generate a video using credits.

    2:26
04

Prompt engineering is the new coding

2 lectures • 17min

  • Understand the LLM generative AI and importance of prompt (LLM input)

    Understand how generative AI is not AGI, and how prompts with large language models produce text, images, and videos, powered by neural networks and data centers.

    11:05

  • Use AI generated prompt for AI

    Learn to craft verbose prompts for AI video generation with diffusion models and large language models, achieving consistent, controllable results.

    6:08
05

Creating custom models, characters, and cameos

3 lectures • 8min

  • Generating stock "talking head" avator

    Explore how CapCut enables AI avatar video creation, enhanced audio, and rapid stitching of stock materials, while noting limitations and the value of a pro subscription.

    4:01

  • Generating your own cameo in Sora

    Keep in mind the cloth you wear at Cameo creation will also be captured and used in the future.

    1:07

  • Generating complex customer cameos

    Create complex cameos by building multiple characters in Sora and OpenArt, then stitch 5–10 second segments into a cohesive storyline using prompts from ChatGPT.

    2:55
06

Trade crafts basic: putting things together

6 lectures • 18min

  • Movie editting: Capcut, iMovies, ClipChimp, etc

    Master three core skills for ai movie making: long-form video editing, ai image generation, and image editing; compare free tools like iMovie, Capcut, and Clipchamp.

    1:58

  • Image generatio with AI - Gemini, OpenAI

    Learn how AI can generate images using Gemini or OpenAI's ChatGPT, craft prompts in a prompt box, and save or download results, while noting that repeated prompts produce different images.

    1:28

  • Image editting basics

    Explore basic image editing using affordable tools like Photoshop and the free Gympie Gympie. Crop photos, remove backgrounds, work with layered formats, and produce outputs, using PowerPoint for quick edits.

    1:24

  • GIMP picture editting demo

    Master gimp picture editing by layering, scaling, and removing the background through selection and invert, showcasing a subway background swap and the contrast with AI image generator workflows.

    9:01

  • Uploading to Youtube and monetize

    Promote your AI movie by uploading to YouTube, targeting the right audience, and using YouTube's AI process to gauge response; submit audition videos as pilots to build momentum.

    2:06

  • Future outlook and Current limitations November 2025

    The best time to start is when no one is ready. Current limitations: 1. Repeatable controlled motion of characters in controlled, repeatable environments. 2. Length of generation limited to 10-15 seconds 3. Heavily rely on prompt input for consistency and details.  Each run gives different results. 4. Still hard to model human faces exactly.  Faces can only be trained through images.  Too few, the model is not complete.  Too many, the model becomes confused. 5. Still not able to have natural human voice, escape for self voice. 6. You can not include certain "celebrities" in your video as cameo, such as Darth Vader. 7. VEO and Seedance are the only ones with face consistency, but the cost is high at 1080 resolution

    1:59