8 papers · 1 filter
Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding
Shuo Gao, Chenhao Zheng, Zhongzheng Ren
Hour-long videos, streaming feeds, and movie-length content are pushing vision-language models (VLMs) past their breaking point -- the standard recipe of encoding each frame into v…
MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction
Jianing Zhang, Chenhao Zheng, Yajun Yang +10
Motion forecasting is central to visual intelligence: agents must anticipate how objects will move in order to plan actions, reason about physical interactions, and synthesize real…
Synthetic Visual Genome 2: Extracting Large-scale Spatio-Temporal Scene Graphs from Videos
Ziqi Gao, Jieyu Zhang, Wisdom Oluchi Ikezogwo +10
We introduce Synthetic Visual Genome 2 (SVG2), a large-scale panoptic video scene graph dataset. SVG2 contains over 636K videos with 6.6M objects, 52.0M attributes, and 6.7M relati…
TrajTok: Learning Trajectory Tokens enables better Video Understanding
Chenhao Zheng, Jieyu Zhang, Jianing Zhang +6
Tokenization in video models, typically through patchification, generates an excessive and redundant number of tokens. This severely limits video efficiency and scalability. While…
Synthetic Object Compositions for Scalable and Accurate Learning in Detection, Segmentation, and Grounding
Weikai Huang, Jieyu Zhang, Taoyang Jia +5
Visual grouping -- operationalized through tasks such as instance segmentation, visual grounding, and object detection -- enables applications ranging from robotic perception to ph…
Synthetic Visual Genome
Jae Sung Park, Zixian Ma, Linjie Li +9
Reasoning over visual relationships-spatial, functional, interactional, social, etc.-is considered to be a fundamental component of human cognition. Yet, despite the major advances…
One Trajectory, One Token: Grounded Video Tokenization via Panoptic Sub-object Trajectory
Chenhao Zheng, Jieyu Zhang, Mohammadreza Salehi +5
Effective video tokenization is critical for scaling transformer models for long videos. Current approaches tokenize videos using space-time patches, leading to excessive tokens an…
Iterated Learning Improves Compositionality in Large Vision-Language Models
Chenhao Zheng, Jieyu Zhang, Aniruddha Kembhavi +1
A fundamental characteristic common to both human vision and natural language is their compositional nature. Yet, despite the performance gains contributed by large vision and lang…