Weakly Supervised Semantic Segmentation with Convolutional Networks

By Pedro H. O. Pinheiro, Ronan Collobert. 2014

54 citations recorded by the source corpus.

We are interested in inferring object segmentation by leveraging only object class information, and by considering only minimal priors on the object segmentation task. This problem could be viewed as a kind of weakly supervised segmentation task, and naturally fits the Multiple Instance Learning (MIL) framework: every training image is known to have (or not) at least one pixel corresponding to the image class label, and the segmentation task can be rewritten as inferring the pixels belonging to the class of the object (given one image, and its object class). We propose a Convolutional Neural Network-based model, which is constrained during training to put more weight on pixels which are important for classifying the image. We show that at test time, the model has learned to discriminate the right pixels well enough, such that it performs very well on an existing segmentation benchmark, by adding only few smoothing priors. Our system is trained using a subset of the Imagenet dataset and the segmentation experiments are performed on the challenging Pascal VOC dataset (with no fine-tuning of the model on Pascal VOC). Our model beats the state of the art results in weakly supervised object segmentation task by a large margin. We also compare the performance of our model with state of the art fully-supervised segmentation approaches.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs (2016, neutral): “[69], [70], [71], achieving significantly better results than weakly-supervised pre-DCNN systems such as [72].”
Learning Deconvolution Network for Semantic Segmentation (2015, background): “On the other hand, [22] performs semantic segmentation based only on image-level annotations in a multiple instance learning framework.”
Weakly-and Semi-Supervised Learning of a Deep Convolutional Network for Semantic Image Segmentation (2015, methodology): “We demonstrate that it is possible to learn a DeepLab-CRF model delivering 62.2% IOU on the PASCAL VOC 2012 test set by training it on a simple foreground/background segmentation of the bounding box annotations.”
Constrained Convolutional Neural Networks for Weakly Supervised Segmentation (2015, methodology): “Recently, there has been a series of work in tackling this problem using convolutional neural networks [23, 25, 26] which we discussed in Section 1.”
Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet (2019, background): “Pinheiro & Collobert (2014) adds explicit labelling of single pixels before the aggregation to an image-level label.”
Adversarial Learning for Semi-supervised Semantic Segmentation (2018, methodology): “The methods in [36] and [35] use Multiple Instance Learning (MIL) to generate latent segmentation label maps for supervised training.”
Weakly Supervised Instance Segmentation Using Class Peak Response (2018, methodology): “External localization network is therefore used to initialize object locations [26, 14, 23], and reﬁning low-resolution CNN planes with pre-generated object segment proposal priors.”
One-Shot Learning for Semantic Segmentation (2017, background): “Weak supervision refers to training from coarse annotations like bounding boxes [9] or image labels [30, 31, 33].”