Pavel Tokmakov

17

Papers

78

Total Citations

Papers (17)

pix2gestalt: Amodal Segmentation by Synthesizing Wholes

Understanding Video Transformers via Universal Concept Discovery

Learning Motion Patterns in Videos

A Structured Model for Action Detection

Discovering Objects That Can Move

Object Discovery From Motion-Guided Tokens

Standing Between Past and Future: Spatio-Temporal Modeling for Multi-Camera 3D Multi-Object Tracking

Tracking Through Containers and Occluders in the Wild

Breaking the "Object" in Video Object Segmentation

Learning Video Object Segmentation With Visual Memory

Towards Latent Attribute Discovery From Triplet Similarities

Learning Compositional Representations for Few-Shot Recognition

Learning To Track With Object Permanence

Zero-1-to-3: Zero-shot One Image to 3D Object

Robust Multi-Object 4D Generation for In-the-wild Videos

TAO: A Large-Scale Benchmark for Tracking Any Object

ReferEverything: Towards Segmenting Everything We Can Speak of in Videos