text logos ChatGPT Image prompts

1 month ago

{ "subject": { "type": "conceptual product design", "main_object": "capsule-shaped container", "contents": "multiple pairs of Adidas Superstar sneakers", "branding": { "logo": "Adidas trefoil and text logo", "placement": "prominently displayed on the top of the capsule" } }, "design_elements": { "form": "giant capsule or pill form with metallic seam", "top_half": { "color": "glossy navy blue", "material": "plastic with reflective highlights" }, "bottom_half": { "material": "transparent acrylic", "contents_visibility": "clear view of the shoes inside" }, "interior": { "shoes": { "type": "Adidas Superstar", "color": "white with black stripes", "arrangement": "tightly packed in orderly layers" } } }, "aesthetic": { "style": "minimalist, product visualization", "color_palette": ["navy blue", "white", "black", "metallic silver", "beige"], "lighting": "studio lighting with soft shadows and high contrast reflections", "mood": "futuristic, brand-focused, surreal marketing concept" }, "camera": { "angle": "frontal, eye-level", "lens": "standard 50mm equivalent", "depth_of_field": "deep, sharp focus throughout" }, "post_processing": { "rendering": "high-resolution CGI or 3D illustration", "finish": "clean, polished, slightly stylized realism" }, "credits": { "creator": "@AD", "platforms": ["Instagram", "TikTok"] } }

4 months ago

A wide image taken with a phone of a glass whiteboard, in a room overlooking the Bay Bridge. The field of view shows a woman writing, sporting a tshirt wiith a large OpenAI logo. The handwriting looks natural and a bit messy, and we see the photographer's reflection. The text reads: (left) "Transfer between Modalities: Suppose we directly model p(text, pixels, sound) [equation] with one big autoregressive transformer. Pros: * image generation augmented with vast world knowledge * next-level text rendering * native in-context learning * unified post-training stack Cons: * varying bit-rate across modalities * compute not adaptive" (Right) "Fixes: * model compressed representations * compose autoregressive prior with a powerful decoder" On the bottom right of the board, she draws a diagram: "tokens -> [transformer] -> [diffusion] -> pixels"