2. Chatgpt et Midjourney pour la reproduction d'image

Pour reproduire le tableau suivant, nous avons utilisé les outils Midjourney et Chatgpt.

Après quelques tweaks, Chatgpt me donne une description adéquate :

"Oil painting in the style of the late Middle Ages or early Renaissance, frontal view, full scene inside a small, compressed hall with marble tiled floor in alternating light and dark geometric patterns and a square wooden ceiling. The lighting is flat and even, with no visible source or shadows. On the right, on a slightly raised platform, an emperor sits upright on a high-backed throne, wearing a thin golden crown with a cross, a long red fur-trimmed robe, holding a scepter in his right hand and touching his chest with his left. Around him stand male courtiers: one with a dark red cap and long robe leaning toward him, another in black observing silently, others forming a second row. At his feet lies a small curled dog facing the center. In the middle of the floor, a kneeling woman in a red robe and light veil presents a round metal platter displaying a symbolic bust — a sculpted male head with closed eyes — turned upward toward the emperor. In front of her stands a small wheeled metal brazier glowing red. To the left, two men watch: one in a red-and-gold patterned outfit and flat black hat extends his hand, palm outward; another in black points toward the scene. On the far right, one man raises both hands in discomfort while another leans against a column, watching silently. Through a Gothic archway in the background, an outdoor scene shows a walled city with towers, men digging a pit, one carrying a shrouded body, one kneeling, another raising a weapon, with red-roofed buildings, a winding road, and trees beyond. The painting is narrative, detailed, and solemn."

ou encore

"/imagine prompt: Oil painting in the style of the late Middle Ages or early Renaissance, frontal view, highly detailed and solemn narrative scene inside a small compressed hall with a marble floor of alternating light and dark geometric patterns and a square wooden coffered ceiling. Flat and even lighting, no visible source or shadows. On the right, on a slightly raised platform, an emperor sits upright on a high-backed throne with a green tapestry backdrop, wearing a thin golden crown with a cross and a long red fur-trimmed robe, holding a scepter in his right hand and touching his chest with his left. Around him stand male courtiers: one with a dark red cap and long robe leaning toward him, another in black observing silently, others forming a second row. At his feet lies a small curled dog facing the center. In the center, a kneeling woman in a red robe and light veil presents a round metal platter displaying a simple stone bust sculpture turned upward toward the emperor. In front of her stands a small wheeled metal brazier glowing red. To the left, two men watch: one in a red-and-gold patterned outfit with a flat black hat extends his hand, palm outward; another in black points toward the scene. On the far right, one man raises both hands in discomfort while another leans against a column, watching silently. Through a Gothic archway in the background, an outdoor scene shows a walled medieval city with towers, where several men work together to dig a large hole, with red-roofed buildings, a winding road, and trees beyond. --ar 2:3 --v 5 --no text"

J'utilise la première sur Midjourney pour générer la même image, ce qui me donne ceci :

Puis sur Chatgpt :

Ce n'était pas parfait. Aussi, c'est vers l'animation vidéo que nous allions nous diriger. Mais avant, nous allions générer un son, afin de véritablement confectionner un clip. La génération sonore est notre prochaine étape.