Large Scale Movie Description Challenge (LSMDC) 2019

Thanks to all participants for making it an exciting challenge and workshop!

Challenge organizers' slides are available here: PPTX with video (107MB) or PDF (2MB).

Tasks

This year we introduce a new challenge, aiming at a more realistic and practical setting of multi-sentence movie description generation. Specifically, movie descriptions are evaluated on sets of 5 clips, rather than on individual clips. When describing sequences of events, it becomes important to distinguish "who is who" in order to provide a coherent and informative narrative. Thus, the challenge will have a focus on identifying characters, rather than predicting generic "SOMEONE"-s in place of all the occurring character names. The challenge is split in three tasks:

• the 1st task still asks to generate descriptions with "SOMEONE"-s but the evaluation is carried out on sets of 5 clips,

• the 2nd task focuses on filling-in the character IDs locally (within a set of 5 clips), while the rest of each description is available,

• the 3rd task combines both description generation and filling-in the local character IDs for the sets of 5 clips; to formally participate in this task is is required to also participate in the 2nd task.

More details are provided here and the corresponding annotations are available for download here.

Winners!

• Workshop winner team is:

  • Youngjae Yu, Jiwan Chung, Jongseok Kim, Heeseung Yun, Gunhee Kim from Seoul National University Vision and Learning Lab, for the 1st place in both Task 1 and 2. Slides: PDF (6 MB)

• Honorable mention goes to the team:

  • Andrew Brown, Samuel Albanie, Yang Liu, Arsha Nagrani, Andrew Zisserman from the Visual Geometry Group at Oxford University, for the 2nd place in Task 2 of the challenge.

Sponsors

We are very happy to have NVIDIA as a sponsor this year!

NVIDIA generously provided a Titan RTX GPU, awarded to the challenge winners (photo above).

Important dates

Challenge submission deadlines

Task 1 "Multi-Sentence Description": October 1st

Task 2 "Fill-in the Characters": October 1st

Task 3 "Multi-Sentence Description with Characters": October 1st

Workshop: Monday, Oct 28th (full day)

Contact

lsmdc2015+19 at gmail.com

Organizers

Anna Rohrbach, Research Scientist at UC Berkeley

Jae Sung Park, Student Assistant at UC Berkeley, PhD student at University of Washington