Ahli PixAI
PixAI
PixAI
Generate
Sign in
Laman Utama
Jana
Model
Recipe
NEW
Studio
Kontes
Mio.2

Conversational creation

Kotak Alat
Guide

User LoRA

SD

Photoreal Emotion Helper

#Happy#Angry#Surprised
Laman UtamaModelMio.2Profil

Alat AI

  • OC Maker
  • AI Anime Generator
  • VTuber Maker
  • Sonic OC Maker
  • Hazbin Hotel OC Maker
  • Anime Character Creator
  • AI Figure Generator
  • Genshin OC Maker
  • All AI Tools

Temui

  • Tag Popular
  • Ranking
  • Pasaran Model
  • Kontes
  • Berita

Tentang

  • Tentang PixAI
  • Guide
  • Cara Menggunakan PixAI
  • Tsubaki.2
  • Kenali Mio
  • Peraturan Kandungan

Harga & Bantuan

  • Keanggotaan
  • Pek Kredit
  • Kontak

Aplikasi Mudah Alih

  • Dapatkan aplikasi PixAI
  • App Store
  • Google Play
© 2026 PixAI
  • Syarat Perkhidmatan
  • Dasar Privasi
  • Dasar Hak Cipta
  • Attributions
  • Undang-undang Transaksi Komersial Tertentu
Photoreal Emotion Helper - AI Model cover image
Dimuat naik pada
3 Ogo 2024, 6:37 PG
Kegunaan
332
Ulasan
Tiada ulasan
Kata Pencetus

exp_real, Happy,

Kebenaran
  • Allow image generation & sharing
  • Benarkan pengguna memuat turun model anda
  • Kegunaan komersial

Keterangan

I'm still shocked at how well this worked. I've spent a year trying out lora intended to give SD 1.5 photoreal models stronger emotions in their facial expressions. I thought training something effective would be significantly harder.

Just use the trigger phrase "exp_real," and one or more of the following: happy, unhappy, angry, worried, shocked, surprised.

("Shocked" and "Surprised" wound up pretty redundant, but you can combine them as another way to boost intensity)

You can get stronger effects with higher weight, 1.0 and higher is often viable. Main side effects appear to be a bias towards brightly lit environments and younger faces.

Full disclosure: This lora was initially intended to be for specific lewd scenarios. It doesn't seem to significantly boost the odds of getting nsfw output, but if you're using nsfw prompts their effects might also be altered (for good or for ill). The training dataset was also completely female, and I have yet to properly test how well it does for men.

EXP_REAL, ANGRY, HAPPY, UNHAPPY, WORRIED, SURPRISED, SHOCKED,

Komen