Prompt templates for ChatGPT object removal and background swaps that don't fall apart
TawohAwa · x · 2026-11-09
Entries 7-8 of TawohAwa's ChatGPT photo-editing prompt series: for object removal, ask the model to realistically reconstruct the area behind with matched lighting, shadows, textures, reflections and perspective; for background swaps, explicitly lock the subject's face, body, outfit and pose while matching light direction, color tones, depth of field and scale. Both are copy-paste ready.
More from Multimodal
- mark_k shares AI-generated fictional megastructure aqueduct image — mark_k · 2026-11-10
- MiniMax H3-Image enters Early Access with 16-reference-image generation and game asset workflows — SarahAnnabels · 2026-11-10
- AI-built 3D desert game world editor demo stuns with "They can build now?" — jasonkneen · 2026-11-10
- Open-source full-duplex voice model Sophon-Realtime v1 beats Grok on tool calling — bdsqlsz · 2026-11-10
- Disney partners with Adobe Firefly to build brand-safe custom AI tools — The Decoder · 2026-11-10
- GPT-image 2.6 'brandshaped typography' prompt turns brand text into logo shapes — aziz4ai · 2026-11-10