top of page

GenAI Portfolio

Brands/Commercial work

Britannia

Project: CGI AI Assistant for Britannia

Goal: Design a CGI holographic "talking head" AI assistant resembling Ravi Shastri to communicate an offer.

Approach:

  1. Image Processing: Used a stock image of Ravi Shastri as the visual base.

  2. Lip Sync Animation: Created realistic lip sync using Runway AI.

  3. Audio Voiceover: Generated the voiceover using Eleven Labs AI to match Ravi Shastri’s style.

  4. Design and Motion Graphics: Designed a futuristic holographic look and applied dynamic motion graphics in After Effects to enhance the visuals.

  5. Tracking and Compositing: Tracked the design with Mocha Pro and composited it all into the final video.

Asset Breakdown

London Dairy

Project: Product Film for London Dairy Ice Cream


Challenges Addressed:

  1. Hero Shot Adjustment: The client needed a wider hero shot, but the footage was initially shot in close-up. Using Adobe Firefly’s generative AI, the background was extended, effectively transforming the close-up into a wide shot.

  2. Product Texture Enhancement: In the selected product window shot, the ice cream appeared melted. Generated a solid ice cream texture to restore the desired look and composited it onto the product.

Reframe Breakdown

Ice Cream scene Breakdown

 AnimateDiff + ComfyUI

AI Model used

protovisionXLHighFidelity3D_releaseV660Bakedvae

Prompt used

((masterpiece, best quality)), Origami young man, folding sculpture, wearing green origami shirt, blue origami jeans, white origami shoes, depth of field, detailed, sharp, 8k resolution, very detailed, cinematic lighting, trending on artstation, hyperdetailed

Output

Exploring GenAI for Production Workflows

Face Replacement (Running locally)

Utilized FaceFusion with the Pinokio AI module on a local machine for face replacement, enabling seamless and realistic face swaps directly in the production environment. This setup offers flexibility and precision in facial modifications while maintaining control over the creative process.

Original (Target)

Shahrukh Khan (Target)

Output Shahrukh Khan

SRK.png

Elon Musk (Target)

Output Elon Musk

Elon_Musk_Royal_Society_crop.jpg

Morphing Animations with ComfyUI | AnimateDiff​​ | Juggernaut Module

Created a morphing animation using AnimateDiff in ComfyUI. Started by generating four base images with Adobe Firefly, then used the Juggernaut module within ComfyUI to transition these images into a seamless, looped image-to-video animation. This approach allowed for fluid transformations and dynamic motion, resulting in a visually captivating animation sequence.

Images generated (Text to image) on Adobe Firefly

Firefly closeup of young adult big eyes cute smiling photorealistic_tech around 46766.jpg
Firefly closeup tight shot of old white men photorealistic_gold ornaments around face 33769.jpg
Firefly closeup tight shot of old skull grunge men photorealistic_with brass crown on head decompose.jpg
Firefly closeup of kid big eyes cute photorealistic_flowers around 8789.jpg

The animation explores the life cycle, beginning with the birth of a child surrounded by nature, symbolizing innocence and growth. As the child matures, they become immersed in technology and the pursuit of wealth. Ultimately, at life’s end, they leave behind all accumulated possessions, decomposing back into nature. The loop then restarts, depicting a newborn surrounded by flowers, representing renewal and the perpetual cycle of life.

Final Output

Image to Video

Project Title: "The City of Tomorrow"

​

Objective:
To create a 15-second AI-generated film showcasing a day in the life of a futuristic eco-city, where nature and technology thrive together.

​

Concept:
The film highlights the seamless harmony between innovative architecture and lush green spaces in a sustainable metropolis. Each scene transitions from morning to night, capturing the vibrant energy of a city designed to balance human life with nature.

Images Generated form Firefly

1.jpg
2.jpg
3.jpg

Image to video from Runway

Text to video

Title: The Growth of a Sapling in Spring

Duration: 5–10 seconds

Objective: To visually capture the journey of a tiny sapling as it grows, symbolizing new beginnings and the beauty of nature’s cycle.

Description:
The video will show a cinematic timelapse of a small sapling breaking through the soil in a lush forest setting during spring. The sapling grows steadily, unfurling its bright green leaves and reaching towards the sunlight as the morning unfolds. The surrounding forest floor is full of vibrant greenery and wildflowers, with subtle motion in the plants as if they are breathing with the sapling. The scene will progress quickly to show the sapling's growth stages, creating a sense of vitality and the promise of new life.

Visual Style:

  • Color Palette: Earthy greens, rich browns, and soft yellows for the morning sunlight

  • Lighting: Natural, warm, and soft, with sunlight filtering through trees

  • Focus: Close-up on the sapling with a shallow depth of field, emphasizing the details of growth

  • Mood: Uplifting, hopeful, and calming, highlighting the beauty of nature

Text to video Minimax

Frame Colorizer using Face Fusion

FaceFusion's Frame Colorizer transforms black-and-white frames into full-color video, utilizing its dataset to produce accurate and realistic colorization. This tool enhances the visual appeal of monochrome footage by automatically adding vibrant colors, making it a powerful option for restoring and modernizing vintage or grayscale media.

Original

AI Colored

Original

AI Colored

Memes IRL Testing

Experimenting with Luma AI’s image-to-video generation to animate static memes, adding movement for a playful and engaging effect. This casual testing explores the potential of AI to bring humor and creativity to life in a fresh format.

2024
bottom of page