We invite researchers and practitioners in computer vision, information retrieval, and human-computer interaction to participate in the 1st International Workshop on Interactive Video Search and Exploration (IViSE 2025), held in conjunction with CVPR 2025.

📅 Date: June 11 – June 15
📍 Location: CVPR 2025 @ Nashville TN
🌐 Website: https://sites.google.com/view/ivise2025

Workshop Overview

The field of video understanding and retrieval has made significant strides in recent years, but current AI-driven methods struggle with long-form video content. IViSE 2025 aims to address this challenge by exploring human-machine teaming approaches for video search and question answering (QA).
This workshop will serve as a venue to compare “fully automated” approaches with “interactive, human-in-the-loop” systems for video retrieval. The focus will be on real-world scenarios involving large-scale video archives, bridging gaps between machine learning, computer vision, and user-centered interactive search strategies.

IViSE 2025 Challenge: Text-Based Video Retrieval and Question Answering

The IViSE 2025 challenge will use the Vimeo Creative Commons Collection (V3C) dataset—comprising 7,475 videos with a total duration of 1,000 hours—to evaluate state-of-the-art methods in:

1️⃣ Fully-Automated Track
Queries are provided in advance.
Participants develop end-to-end AI-based video retrieval and QA systems.
Solutions are evaluated on accuracy and retrieval effectiveness.

2️⃣ Interactive Track
Queries are presented during the workshop.
Teams must solve tasks in real-time within five minutes, leveraging human-machine collaboration.

Inspired by the Video Browser Showdown and Lifelog Search Challenge:
🔹 Known-Item Search (KIS): Given a textual description, retrieve the correct video segment and time interval.
🔹 Question Answering (QA): Answer a question based on a textual description of a video.

Both tracks will use standardized evaluation metrics and build upon existing benchmarks such as TRECVID & DVU for automated search, and Video Browser Showdown for interactive retrieval.

Why Participate?
Advance the field of long-form video retrieval and exploration.
Benchmark your methods against cutting-edge AI and interactive systems.
Engage with leading researchers in computer vision, video retrieval, and interactive search.
Compete in a structured challenge format with an emphasis on both fully automated & interactive approaches and also in a high-impact challenge at CVPR 2025.

Important Dates

📅 Feb 3, 2025 – Fully-Automated Track queries released
📅 Mar 24, 2025 – Submission deadline for Fully-Automated Track results & workshop papers
📅 Apr 3, 2025 – Reviews and challenge scores released
📅 Apr 7, 2025 – Camera-ready deadline

We look forward to your participation in IViSE 2025 as we explore the next frontier in interactive video search and long-form video understanding