Video inpainting can't yet remove camera rigs: one user's failed VACE workflow attempt
Medium_Chemist_4032 · reddit · 2026-09-23
A user tried to erase a camera rig from real footage via video inpainting. Using a VACE workflow with masks, they couldn't get a single reliable generation, and found no solid partial-inpainting solution—results were full of artifacts. The post is a request for workflows, but also a data point that current video models still struggle with this task.
More from Multimodal
- Google launches Gemini 3.8 Flash TTS and Flash-Lite TTS, its most expressive audio models — patloeber · 2026-09-23
- Dev open-sources a local Mac studio for testing Qwen-Image-2.1 image generation — janishar · 2026-09-23
- Adobe completes acquisition of Emmy-winning AI enhancer Topaz Labs — chrisfirst · 2026-09-23
- Google DeepMind debuts Gemini 3.8 text-to-speech model — Google DeepMind · 2026-09-23
- Google DeepMind launches Gemini 3.8 TTS with 30-second voice cloning and consent checks — Google DeepMind · 2026-09-23
- Recraft launches V4.1 Flash, generating images in 1.3 seconds as fastest image model — aziz4ai · 2026-09-23