Workshop on Video-Language Models
About the workshop
The growing relevance of video-language models in both academia and industry highlights the necessity for a dedicated workshop to address the unique challenges and opportunities this field presents. This workshop is designed to accelerate the development and practical application of video foundation models, which are crucial for interpreting and utilizing the extensive amounts of video data that make up a significant portion of global data. These models are increasingly vital for a range of applications, from video search and content creation to surveillance and robotics.
Our workshop will tackle four primary challenges:
- First, the scarcity of high-quality, annotated video data is a significant barrier to progress. Unlike text and image data, which are abundant and often come with high-quality annotations, video data typically lacks such detailed annotations, limiting the development of advanced models.
- Second, the sheer volume of video data demands significant advancements in data processing techniques.. Modern video models must process hundreds to thousands of frames per video, with each frame requiring detailed analysis. Efficient processing methods are needed to handle this scale while maintaining detailed information capture.
- Third, the multimodal nature of video data requires sophisticated model designs that can integrate audio, visual, temporal, and textual data in a cohesive manner.
- Last but not least, the community still lacks robust video-language alignment benchmarks, which makes it hard to evaluate and compare the capabilities of video-language models.
Featuring organizers from leading AI institutions
This workshop will serve as a platform for sharing knowledge, fostering collaborations, and setting future research directions in this essential and rapidly advancing field.
What the workshop offers
Resources such as access to embeddings for large-scale public video datasets and benchmarks for video-language alignment (See Video Embeddings and Benchmark Section).
Opportunity to explore the ethical implications of video foundation models, focusing on safety, reliability, and responsibility.
A series of expert talks, panel discussions, and collaborative sessions to discuss current advancements and tackle existing challenges.
Resources such as access to embeddings for large-scale public video datasets and benchmarks for video-language alignment (See Video Embeddings and Benchmark Section).
Opportunity to explore the ethical implications of video foundation models, focusing on safety, reliability, and responsibility.
A series of expert talks, panel discussions, and collaborative sessions to discuss current advancements and tackle existing challenges.
Panelists: TBD
Call for papers
SubmitWe will invite researchers to submit papers focusing on, but not limited to, the following topics related to video-language models:
- Video-text alignment and multimodal understanding
- Text-to-video generation and editing using natural language prompts
- Video-to-text generation, including video captioning and description
- Temporal reasoning and event understanding in video language models
- Cross-modal retrieval between video and text
- Video question answering and visual dialogue systems
- Long-form video understanding and summarization
- Ethical considerations and bias mitigation in video AI
- Benchmarks and evaluation metrics for video-language tasks
- Multimodal fusion techniques for video, audio, and text
- Short track: Up to 3 pages. Submission of abstract that shows early, novel ideas related to the workshop
- Long track: Up to 9 pages excluding the references. Submission of papers relevant to our workshop, or accepted at other conferences.
- Video question answering and visual dialogue systems
- Long-form video understanding and summarization
- Ethical considerations and bias mitigation in video AI
- Benchmarks and evaluation metrics for video-language tasks
- Multimodal fusion techniques for video, audio, and text
Tentative important dates
- Abstract Submission Deadline:
September 14, 2024 - Paper Submission Deadline:
September 17, 2024 - Review Bidding Period:
September 17 - September 20, 2024 - Review Deadline:
October 11, 2024 - Acceptance/Rejection Notification Date:
October 14, 2024 - Workshop Date:
December 14 or 15, 2024
Awards
Among exceptional research papers with high review scores, we will select one best paper award and two runner-ups.
We will invite researchers to submit papers focusing on, but not limited to, the following topics related to video-language models:
- Video-text alignment and multimodal understanding
- Text-to-video generation and editing using natural language prompts
- Video-to-text generation, including video captioning and description
- Temporal reasoning and event understanding in video language models
- Cross-modal retrieval between video and text
- Video question answering and visual dialogue systems
- Long-form video understanding and summarization
- Ethical considerations and bias mitigation in video AI
- Benchmarks and evaluation metrics for video-language tasks
- Multimodal fusion techniques for video, audio, and text
Short, archival track: Submission of abstract that shows early, novel ideas related to the workshop (up to 3 pages)
Long, non-archival track: Submission of papers relevant to our workshop, or accepted at other conferences.
Tentative Important Dates
- Abstract Submission Deadline:
September 10, 2024 - Paper Submission Deadline:
September 13, 2024 - Review Bidding Period:
September 13 - September 17, 2024 - Review Deadline:
October 11, 2024 - Acceptance/Rejection Notification Date:
October 14, 2024 - Workshop Date:
December 14 or 15, 2024
Awards
Among exceptional research papers with high review scores, we will select one best paper award and two runner-ups.
Video embeddings and benchmark: TBD
Processing videos and obtaining their embeddings is crucial for creating a powerful video-language model, but it is also prone to heavy computations and privacy issues. We will release video embeddings and industry-grade evaluation benchmarks to facilitate the research