Learn to Propagate Reliably on Noisy Affinity Graphs
This addresses the challenge of reducing labeling costs in visual recognition by enabling more reliable propagation on large-scale, noisy datasets, representing an incremental improvement over existing methods.
The paper tackles the problem of unreliable label propagation on noisy affinity graphs, proposing a framework that improves overall accuracies, with experiments on ImageNet and Ms-Celeb-1M showing significant gains, especially in noisy conditions.
Recent works have shown that exploiting unlabeled data through label propagation can substantially reduce the labeling cost, which has been a critical issue in developing visual recognition models. Yet, how to propagate labels reliably, especially on a dataset with unknown outliers, remains an open question. Conventional methods such as linear diffusion lack the capability of handling complex graph structures and may perform poorly when the seeds are sparse. Latest methods based on graph neural networks would face difficulties on performance drop as they scale out to noisy graphs. To overcome these difficulties, we propose a new framework that allows labels to be propagated reliably on large-scale real-world data. This framework incorporates (1) a local graph neural network to predict accurately on varying local structures while maintaining high scalability, and (2) a confidence-based path scheduler that identifies outliers and moves forward the propagation frontier in a prudent way. Experiments on both ImageNet and Ms-Celeb-1M show that our confidence guided framework can significantly improve the overall accuracies of the propagated labels, especially when the graph is very noisy.