Can Image-Level Labels Replace Pixel-Level Labels for Image Parsing

Abstract

This paper presents a weakly supervised sparse learning approach to the problem of noisily tagged image parsing, or segmenting all the objects within a noisily tagged image and identifying their categories (i.e. tags). Different from the traditional image parsing that takes pixel-level labels as strong supervisory information, our noisily tagged image parsing is provided with noisy tags of all the images (i.e. image-level labels), which is a natural setting for social image collections (e.g. Flickr). By oversegmenting all the images into regions, we formulate noisily tagged image parsing as a weakly supervised sparse learning problem over all the regions, where the initial labels of each region are inferred from image-level labels. Furthermore, we develop an efficient algorithm to solve such weakly supervised sparse learning problem. The experimental results on two benchmark datasets show the effectiveness of our approach. More notably, the reported surprising results shed some light on answering the question: can image-level labels replace pixel-level labels (hard to access) as supervisory information for image parsing.

References in corpus (4)

• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• Fully Convolutional Multi-Class Multiple Instance Learning
• Smoothing Proximal Gradient Method for General Structured Sparse Learning
• Robust Subjective Visual Property Prediction from Crowdsourced Pairwise Labels

Cited by in corpus (19)

• Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey
• Deep learning approach to Fourier ptychographic microscopy
• Deep Tone Mapping Operator for High Dynamic Range Images
• Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective
• Structure Label Prediction Using Similarity-Based Retrieval and Weakly-Supervised Label Mapping
• Guided Anisotropic Diffusion and Iterative Learning for Weakly Supervised Change Detection
• Effect of Annotation Errors on Drone Detection with YOLOv3
• CEREALS - Cost-Effective REgion-based Active Learning for Semantic Segmentation
• Weakly Supervised Change Detection Using Guided Anisotropic Difusion
• Segmentation-Aware Image Denoising without Knowing True Segmentation
• PCAMs: Weakly Supervised Semantic Segmentation Using Point Supervision
• Multi-Modal Multi-Scale Deep Learning for Large-Scale Image Annotation
• How to Efficiently Annotate Images for Best-Performing Deep Learning Based Segmentation Models: An Empirical Study with Weak and Noisy Annotations and Segment Anything Model
• Deep Robust Subjective Visual Property Prediction in Crowdsourcing
• Superpixel Segmentation Using Gaussian Mixture Model
• Deep Learning for Earth Image Segmentation based on Imperfect Polyline Labels with Annotation Errors
• Brain Metastasis Segmentation Network Trained with Robustness to Annotations with Multiple False Negatives
• DASNet: Reducing Pixel-level Annotations for Instance and Semantic Segmentation
• Joint Optical Neuroimaging Denoising with Semantic Tasks