İçeriğe geç
YENİPazar yeri ve trig8n Agent: geliştirdiğinizi satın; planlar ve değişiklikler siz görmeden önce test edilsin. Yeniliklere bak

Kütüphane / Görüntü, Video ve Ses

Görüntü, Video ve Ses

GPT-4 ve Dumpling AI ile Reklam Görseli Varyasyonları Üretme

Marka bilgileri ve bir reklam görseli gönderildiğinde GPT-4 görseli tanımlar ve Dumpling AI, Google Drive'a kaydedilen on görsel varyasyonu oluşturur.

Ne yapar

### 🖼️ Ad Image Variation Generator (Using GPT-4 + Dumpling AI) This workflow creates 10 subtle creative variations of a reference ad image to test performance across visual styles, lighting, background, and tone — while preserving the product's framing and subject. --- ### 🔧 How It Works 1. User submits brand name, website, and reference ad image via a form. 2. The image is uploaded to Google Drive. 3. GPT-4o analyzes the image’s visual style (composition, subject, lighting). 4. GPT-4 analyzes the brand website to understand overall visual identity. 5. A LangChain AI Agent uses both to generate 10 variation prompts. 6. Each prompt is passed to Dumpling AI to generate a new ad image. 7. All image URLs are logged in Google Sheets.

Adımlar

  1. Submit Brand Info + Image
  2. Upload Ad Image to Google Drive
  3. Download Ad Image for Analysis
  4. Describe Visual Style of Image
  5. Analyze Brand Website Style
  6. LangChain Agent: Generate Variation Prompts
  7. Parse Prompts into JSON Array
  8. Split: One Prompt per Item
  9. Download Base Image for Each Variation
  10. Loop: Process Image Variations
  11. Dumpling AI: Generate Image Variation
  12. Log Image Variation URLs to Google Sheets
  13. GPT-4o (Connected to LangChain Agent)

Benzer otomasyonlar