Creating anthropomorphic animals with Gemini and HIggsfield
The speaker describes a workflow for creating anthropomorphic animal exercise videos using two AI tools. Animals are generated in Gemini, while Higgsfield is used to merge those still images with self-filmed exercise footage. The result is an animated anthropomorphic animal appearing to perform the exercises.
Summary
The speaker outlines a creative AI-assisted video production process developed to produce anthropomorphic animal exercise content. The workflow begins in Gemini, where the speaker generates still images of animals. A key discovery was that the starting position or pose of the animals in these images was critically important to the final output quality.
The speaker's husband, who works extensively in AI, introduced her to Higgsfield, a platform with a specific capability to merge or blend a still image with a motion video. This feature allows the subject depicted in the still image to appear as though it is performing the actions shown in the video.
To complete the workflow, the speaker films herself performing exercises, then uses Higgsfield to combine the Gemini-generated animal image with her exercise footage. The end result is a video of an anthropomorphic animal — referred to as 'Bryce' — performing the exercises, effectively replacing or overlaying the speaker's movements onto the animal character.
Key Insights
- The speaker found that the starting pose or position of the animal in the Gemini-generated still image was critically important to the success of the final video output.
- The speaker uses Gemini specifically for generating still images of anthropomorphic animals, not for video creation.
- The speaker's husband, who works in AI, was the one who introduced her to Higgsfield, suggesting the workflow was discovered through personal collaboration rather than independent research.
- Higgsfield's key capability described is merging a still image with a motion video so that the subject in the image appears to perform the actions shown in the video.
- The speaker films herself performing the exercises and uses that self-recorded footage as the motion source that gets transferred onto the AI-generated animal character.
Topics
Transcript
[0:00] I ended up creating the animals in Gemini. I came to learn that the starting position for the animals was really key. Gemini wasn't doing the videos, but I was making the animals there, and my husband does a lot of work in AI, and he introduced me to Higgs Field, and one of their capabilities is to merge, mash, intertwine a still image with a motion video to have what's in [0:30] the image doing what's in the video. I make the animal in Gemini. I film myself doing the exercises, and then I mash up the anthropomorphic animal with Bryce exercising to create the video.
Full transcript available for MurmurCast members
Sign Up to AccessMore from How I AI
GPT-5.6's video editing via Codex is genuinely one of my favorite new workflows
A product manager describes using GPT-5.6 with Codex to automate video editing for social media clips. By simply dragging a file and providing natural language instructions, the AI generated five polished, fast-paced hype videos from a long conference talk recording, dramatically reducing the time-intensive manual clipping process.
Theoretically Intelligent vs. Practically Effective: Why GPT-5.6 Sol Beats Fable for Product Work
An executive contrasts two AI models (Fable and Soul/GPT-5.6 Sol), arguing that Soul is superior for product work because it prioritizes practical effectiveness over theoretical intelligence. The speaker values the ability to ship products to customers and understand end-user goals over theoretical sophistication.
Build a harness when the same workflow needs the same setup and the same outcomes, every time
Building a harness for repetitive workflows allows you to be more prescriptive about job execution, resulting in greater efficiency, consistency, and better outcomes. Rather than explaining requirements to an AI agent each time, a harness lets you use a simpler interface like pasting a link while the agent already understands the intended task.
GPT 5.6-Sol vs. Claude Fable: Why OpenAI’s new model crushes my benchmark
Claire Vo compares OpenAI's new GPT 5.6 models (Soul, Terra, Luna) against Claude's Fable using her custom "How I AI" benchmark, finding that GPT 5.6 Soul excels at practical product work, prototyping, and natural communication, while Fable is theoretically intelligent but pedantic and difficult to collaborate with.
Context offloading is an underrated AI use case
The speaker highlights context offloading as an underrated AI use case, where AI serves as a safety net for routine cognitive tasks like email management and personal finances. Rather than adding new capabilities, AI reduces anxiety about missing important information or making mistakes by handling monitoring tasks, thereby freeing up mental bandwidth.