Built independently by an author, for readers. Read the story and support ChapterPal

keyword

video copy detection models

Video copy detection models are machine learning systems designed to identify whether a query video or video segment is an exact duplicate, modified version, or partial reuse of an existing reference video. Unlike general video classification or semantic retrieval systems that group videos by conceptual category or theme, video copy detection models focus on instance-level matching by extracting robust spatial and temporal visual fingerprints. These models are engineered to remain invariant to common post-processing transformations and digital manipulations, such as resolution scaling, cropping, frame rate alterations, compression, color grading, picture-in-picture insertions, and temporal reordering. As a result, they are extensively utilized for copyright enforcement, intellectual property tracking, automated content moderation, dataset deduplication, and multimedia asset management.

1 item

VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models

VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models

Wenhao Wang, Yi Yang

OrganizationsUniversity of Technology SydneyZhejiang University

Why you should read this

Introduces VidProM, the first large-scale prompt-gallery dataset containing 1.67 million real-user text-to-video prompts paired with 6.69 million synthesized videos from four diffusion models to advance research in video prompt engineering, efficient generation, and synthetic video detection.

The arrival of Sora marks a new era for text-to-video diffusion models, bringing significant advancements in video generation and potential applications. However, Sora, along with other text-to-video diffusion models, is highly reliant on prompts, and there is no publicly available dataset that features a study of text-to-video prompts. In this paper, we introduce VidProM, the first large-scale dataset comprising 1.67 Million unique text-to-Video Prompts from real users. Additionally, this dataset includes 6.69 million videos generated by four state-of-the-art diffusion models, alongside some related data. We initially discuss the curation of this large-scale dataset, a process that is both time-consuming and costly. Subsequently, we underscore the need for a new prompt dataset specifically designed for text-to-video generation by illustrating how VidProM differs from DiffusionDB, a large-scale prompt-gallery dataset for image generation. Our extensive and diverse dataset also opens up many exciting new research areas. For instance, we suggest

Added

2026-09-26