Meta-Learning Deep Visual Words for Fast Video Object Segmentation

Personal robots and driverless cars need to be able to operate in novel environments and thus quickly and efficiently learn to recognise new object classes. We address this problem by considering the task of video object segmentation. Previous accurate methods for this task finetune a model using the first annotated frame, and/or use additional inputs such as optical flow and complex post-processing. In contrast, we develop a fast, causal algorithm that requires no finetuning, auxiliary inputs or post-processing, and segments a variable number of objects in a single forward-pass. We represent an object with clusters, or "visual words", in the embedding space, which correspond to object parts in the image space. This allows us to robustly match to the reference objects throughout the video, because although the global appearance of an object changes as it undergoes occlusions and deformations, the appearance of more local parts may stay consistent. We learn these visual words in an unsupervised manner, using meta-learning to ensure that our training objective matches our inference procedure. We achieve comparable accuracy to finetuning based methods (whilst being 1 to 2 orders of magnitude faster), and state-of-the-art in terms of speed/accuracy trade-offs on four video segmentation datasets. Code is available at https://github.com/harkiratbehl/MetaVOS.

Authors

• Harkirat Singh Behl
• Mohammad Najafi
• Anurag Arnab
• Philip H. S. Torr

Keywords

• Training
• Visualization
• Object segmentation
• Task analysis
• Optical flow
• Intelligent robots
• Strain
• Video Object Segmentation
• Fine-tuned
• Object Classification
• Latent Space
• Number Of Objects
• Image Space
• Self-driving
• Object Parts
• Inference Procedure
• Local Parts
• Unsupervised Manner
• Video Segments
• Bimodal
• Classification Of Samples
• Bounding Box
• Multiple Objects
• Video Frames
• Single Vector
• Cluster Centroids
• Pixel In Frame
• Video Object
• Support Set
• Metric Learning
• Words In The Lexicon
• Query Set
• Nearest Neighbor Search
• Prototypical Network
• Pareto Front
• Triplet Loss