A group of researchers working for AI Chipmaker Nvidia, the third largest company in the world by market capitalization, in collaboration with students graduates from Stanford, UC San Diego, UC Berkeley and UT Austin, formed an AI system on 81 Tom & Jerry classic theatrical currents to see if they could create long animated sequences.
In appearance, the purpose of the exercise was to generate a longer and more coherent AI video while overcoming the challenge of “self-attenuated”, a technical principle that currently performs the prohibition of video generated by the video generated by interstation. All the current public APIs of the AI Videos generation are still limited by the duration: since last month, Sora d’Openai has a maximum length of 20 seconds, Meta’s Moviegen is 16 seconds, the Ray 2 of Luma is 10 seconds and the Veo 2 of Google is eight seconds.
The research team (which includes Jiarui Xu, Shihao Han, Ka Chun Cheung, Jan Kautz, Yejin Choi, Yu Sun and Xialong Wang) HERE as well as in a paper (Download PDF) which was published yesterday. They wrote: “The videos tell complex stories with coherent scenes made up of dynamic movement. Each video is produced directly by the model in a single photo, without edition, sewing or post-treatment. Each story is newly created.” (In her article, Nvidia does not recognize the permission of Warner Bros. Discovery to use her characters and her films protected by copyright for such an experience.)
You will find below one of the videos created by Nvidia et al. This shows that Tom works as office employee in the World Trade Center:
At first glance, he looks impressive, but watch for a few seconds and he quickly disintegrates with illogical performances, strange objects, incoherent connections and only the slightest suggestion of coherent narration. Although it is an experience and should be considered in the context of the objectives he has decided to achieve, he only shows that AI is still several years from the generation of an entire animated film alone. It is also notable that the quantity of text even caused to achieve it. Here is the full prompt that was necessary to generate the video above:
The rounds of the World Trade Center stand up against a light blue and lively blue sky. The streets light up with pedestrians in costumes and the yellow taxis move slowly through strong circulation. Sun light is strongly reflected from glass windows from neighboring buildings. Tom, the blue gray cat, quickly walks along the large gray sidewalk with a single black case in hand.
Inside the World Trade Center lobby, expansive marble floors reflect hot golden with gold. Gray marble pillars and brass accessories highlight the elegant entrance, as well as a brass elevator door. A uniform porter carrying a dark navy blue suit stands behind a polished wooden counter. Tom, the blue gray cat, has a single black case in its hand slowly presses the elevator button on the left side of the elevator and waits. Tom’s left hand is empty.
Under the large oak desk, a single intact black cable takes place carefully along the beige carpet to a silver desktop tower. A clean white wall socket safely contains a single candle. Jerry, the brown mouse, thoroughly chews the black cable. Tiny and barely perceptible sparks steal as it bites the cable.
A shiny corner desk includes floor windows on the ceiling overlooking the large New York horizon line under a sunny blue sky. The simple cream -colored walls complete a large executive office in shiny oak. In the center of the office, in front of Tom, there is a silver computer instructor and a silver keyboard, a silver pen and a black leather desk chair. Tom, the blue-gray cat, now with an upset face, leans on the floor to check under the desk.
Under the large oak desk, a single black cable that has exposed copper wire in one place, works carefully along the beige carpet to a silver desktop tower. A clean white wall socket safely contains a single candle. Jerry, the brunette mouse, stands near the black minced cable while Tom, the blue-gray cat, has his head in the angry and surprised scene. Jerry hurry out of Tom out of the scene, while Tom takes off his head from the scene.
A shiny corner office with natural cream -colored walls and a brown oak door is to the right of the scene. There is a small mouse slightly to the left of the door with the plinth. Tom, the blue gray cat, runs towards the mouse hole and slams his body in the wall. Tom stunned with small caricatural stars turning around his head, which indicates dizziness.
A shiny corner office with natural cream -colored walls and a brown oak door is to the right of the scene. There is a small mouse slightly to the left of the door with the plinth. Tom, the Blue-Gris cat, was stunned. Tom, the blue-gray cat, checks the small watch on its right wrist. Tom, the blue-gray cat, reads time on his watch and notes that he is late for a meeting. In a rush, Tom, the Blue-Gris cat, gets back, is no longer amazed, opens the oak door to the right of the scene and goes through the door.
In the large conference room, ordinary cream walls surround a long wooden table positioned in the center. A small black wall clock is positioned at the top right of the screen. There are several executives who are seated in black office chairs are carefully arranged around the table. The leaders are quietly concentrated on Spike, the gray bulldog, which is held on a whiteboard at the front of the room, while Tom, the blue-gray cat, is held to the left of Spike. Spike, the gray bulldog, point with anger on the small black wall clock at the top right of the screen while maintaining aggressive visual contact with Tom, the blue-gray cat. Spike, the gray bulldog, reprimands Tom loudly. Tom, now with a frightened expression, covers his ears because of the cries and makes an excuse gesture.
In the large conference room, ordinary cream walls surround a long wooden table positioned in the center. A small black wall clock is positioned at the top right of the screen. There are several executives who are seated in black office chairs are carefully arranged around the table. The leaders are quietly concentrated on Spike, the gray bulldog, which is held on a whiteboard at the front of the room, while Tom, the blue-gray cat, is held to the left of Spike. Spike, the gray bulldog, point with anger and holds its pointing to the left while looking at Tom. Tom sulks then walks sadly to the left of the scene.
In the comfortable interior of the mouse hole with rounded walls, wooden furniture and warm and soft lighting, Jerry, the brown mouse, continues to laugh happily while the scene gradually fades in black.