FEELVOS: Fast End-to-End Embedding Learning for Video Object Segmentation

Abstract

Many of the recent successful methods for video object segmentation (VOS) are overly complicated, heavily rely on fine-tuning on the first frame, and/or are slow, and are hence of limited practical use. In this work, we propose FEELVOS as a simple and fast method which does not rely on fine-tuning. In order to segment a video, for each frame FEELVOS uses a semantic pixel-wise embedding together with a global and a local matching mechanism to transfer information from the first frame and from the previous frame of the video to the current frame. In contrast to previous work, our embedding is only used as an internal guidance of a convolutional network. Our novel dynamic segmentation head allows us to train the network, including the embedding, end-to-end for the multiple object segmentation task with a cross entropy loss. We achieve a new state of the art in video object segmentation without fine-tuning with a J&F measure of 71.5% on the DAVIS 2017 validation set. We make our code and models available at https://github.com/tensorflow/models/tree/master/research/feelvos.

CVPR 2019 camera-ready version

References in corpus (8)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Rethinking Atrous Convolution for Semantic Image Segmentation
• DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
• The 2018 DAVIS Challenge on Video Object Segmentation
• Semantic Instance Segmentation via Deep Metric Learning
• Bridging Category-level and Instance-level Semantic Image Segmentation
• Learning Video Object Segmentation from Static Images

Cited by in corpus (13)

• Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentation
• Do Different Tracking Tasks Require Different Appearance Models?
• Learning Video Object Segmentation from Unlabeled Videos
• Dynamic Mini-batch SGD for Elastic Distributed Training: Learning in the Limbo of Resources
• Efficient Regional Memory Network for Video Object Segmentation
• Spatial Feature Calibration and Temporal Fusion for Effective One-stage Video Instance Segmentation
• Self-supervised Video Object Segmentation
• A Transductive Approach for Video Object Segmentation
• Crossover Learning for Fast Online Video Instance Segmentation
• Contextual Guided Segmentation Framework for Semi-supervised Video Instance Segmentation
• Efficient Full Image Interactive Segmentation by Leveraging Within-image Appearance Similarity
• Target-Aware Object Discovery and Association for Unsupervised Video Multi-Object Segmentation
• An Empirical Study of Propagation-based Methods for Video Object Segmentation