Yunpeng Chen

STC: A Simple to Complex Framework for Weakly-Supervised Semantic Segmentation

Recently, significant improvement has been made on semantic object segmentation due to the development of deep convolutional neural networks (DCNNs). Training such a DCNN usually relies on a large number of images with pixel-level segmentation masks, and annotating these images is very costly in terms of both finance and human effort. In this paper, we propose a simple to complex (STC) framework in which only image-level annotations are utilized to learn DCNNs for semantic segmentation. Specifically, we first train an initial segmentation network called Initial-DCNN with the saliency maps of simple images (i.e., those with a single category of major object(s) and clean background). These saliency maps can be automatically obtained by existing bottom-up salient object detection techniques, where no supervision information is needed. Then, a better network called Enhanced-DCNN is learned with supervision from the predicted segmentation masks of simple images based on the Initial-DCNN as well as the image-level annotations. Finally, more pixel-level segmentation masks of complex images (two or more categories of objects with cluttered background), which are inferred by using Enhanced-DCNN and image-level annotations, are utilized as the supervision information to learn the Powerful-DCNN for semantic segmentation. Our method utilizes 40K simple images from Flickr.com and 10K complex images from PASCAL VOC for step-wisely boosting the segmentation network. Extensive experimental results on PASCAL VOC 2012 segmentation benchmark well demonstrate the superiority of the proposed STC framework compared with other state-of-the-arts.Recently, significant improvement has been made on semantic object segmentation due to the development of deep convolutional neural networks (DCNNs). Training such a DCNN usually relies on a large number of images with pixel-level segmentation masks, and annotating these images is very costly in terms of both finance and human effort. In this paper, we propose a simple to complex (STC) framework in which only image-level annotations are utilized to learn DCNNs for semantic segmentation. Specifically, we first train an initial segmentation network called Initial-DCNN with the saliency maps of simple images (i.e., those with a single category of major object(s) and clean background). These saliency maps can be automatically obtained by existing bottom-up salient object detection techniques, where no supervision information is needed. Then, a better network called Enhanced-DCNN is learned with supervision from the predicted segmentation masks of simple images based on the Initial-DCNN as well as the image-level annotations. Finally, more pixel-level segmentation masks of complex images (two or more categories of objects with cluttered background), which are inferred by using Enhanced-DCNN and image-level annotations, are utilized as the supervision information to learn the Powerful-DCNN for semantic segmentation. Our method utilizes 40K simple images from Flickr.com and 10K complex images from PASCAL VOC for step-wisely boosting the segmentation network. Extensive experimental results on PASCAL VOC 2012 segmentation benchmark well demonstrate the superiority of the proposed STC framework compared with other state-of-the-arts.

Learning to segment with image-level annotations

Recently, deep convolutional neural networks (DCNNs) have significantly promoted the development of semantic image segmentation. However, previous works on learning the segmentation network often rely on a large number of ground-truths with pixel-level annotations, which usually require considerable human effort. In this paper, we explore a more challenging problem by learning to segment under image-level annotations. Specifically, our framework consists of two components. First, reliable hypotheses based localization maps are generated by incorporating the hypotheses-aware classification and cross-image contextual refinement. Second, the segmentation network can be trained in a supervised manner by these generated localization maps. We explore two network training strategies for achieving good segmentation performance. For the first strategy, a novel multi-label cross-entropy loss is proposed to train the network by directly using multiple localization maps for all classes, where each pixel contributes to each class with different weights. For the second strategy, the rough segmentation mask can be inferred from the localization maps, and then the network is optimized based on the single-label cross-entropy loss with the produced masks. We evaluate our methods on the PASCAL VOC 2012 segmentation benchmark. Extensive experimental results demonstrate the effectiveness of the proposed methods compared with the state-of-the-arts. HighlightsLocalization map generation is proposed by using the hypothesis-based classification.A novel multi-label loss is proposed to train the network based on localization maps.An effective method is proposed to predict the rough mask of the given training image.Our methods achieve new state-of-the-art results on PASCAL VOC 2012 benchmark.

Sharing Residual Units Through Collective Tensor Factorization To Improve Deep Neural Networks

Residual units are wildly used for alleviating optimization difficulties when building deep neural networks. However, the performance gain does not well compensate the model size increase, indicating low parameter efficiency in these residual units. In this work, we first revisit the residual function in several variations of residual units and demonstrate that these residual functions can actually be explained with a unified framework based on generalized block term decomposition. Then, based on the new explanation, we propose a new architecture, Collective Residual Unit (CRU), which enhances the parameter efficiency of deep neural networks through collective tensor factorization. CRU enables knowledge sharing across different residual units using shared factors. Experimental results show that our proposed CRU Network demonstrates outstanding parameter efficiency, achieving comparable classification performance to ResNet-200 with the model size of ResNet-50. By building a deeper network using CRU, we can achieve state-of-the-art single model classification accuracy on ImageNet-1k and Places365-Standard benchmark datasets. (Code and trained models are available on GitHub)The residual unit and its variations are wildly used in building very deep neural networks for alleviating optimization difficulty. In this work, we revisit the standard residual function as well as its several successful variants and propose a unified framework based on tensor Block Term Decomposition (BTD) to explain these apparently different residual functions from the tensor decomposition view. With the BTD framework, we further propose a novel basic network architecture, named the Collective Residual Unit (CRU). CRU further enhances parameter efficiency of deep residual neural networks by sharing core factors derived from collective tensor factorization over the involved residual units. It enables efficient knowledge sharing across multiple residual units, reduces the number of model parameters, lowers the risk of over-fitting, and provides better generalization ability. Extensive experimental results show that our proposed CRU network brings outstanding parameter efficiency—it achieves comparable classification performance with ResNet-200 while using a model size as small as ResNet-50 on the ImageNet-1k and Places365-Standard benchmark datasets.

Learning to Segment Human by Watching YouTube

An intuition on human segmentation is that when a human is moving in a video, the video-context (e.g., appearance and motion clues) may potentially infer reasonable mask information for the whole human body. Inspired by this, based on popular deep convolutional neural networks (CNN), we explore a very-weakly supervised learning framework for human segmentation task, where only an imperfect human detector is available along with massive weakly-labeled YouTube videos. In our solution, the video-context guided human mask inference and CNN based segmentation network learning iterate to mutually enhance each other until no further improvement gains. In the first step, each video is decomposed into supervoxels by the unsupervised video segmentation. The superpixels within the supervoxels are then classified as human or non-human by graph optimization with unary energies from the imperfect human detection results and the predicted confidence maps by the CNN trained in the previous iteration. In the second step, the video-context derived human masks are used as direct labels to train CNN. Extensive experiments on the challenging PASCAL VOC 2012 semantic segmentation benchmark demonstrate that the proposed framework has already achieved superior results than all previous weakly-supervised methods with object class or bounding box annotations. In addition, by augmenting with the annotated masks from PASCAL VOC 2012, our method reaches a new state-of-the-art performance on the human segmentation task.An intuition on human segmentation is that when a human is moving in a video, the video-context (e.g., appearance and motion clues) may potentially infer reasonable mask information for the whole human body. Inspired by this, based on popular deep convolutional neural networks (CNN), we explore a very-weakly supervised learning framework for human segmentation task, where only an imperfect human detector is available along with massive weakly-labeled YouTube videos. In our solution, the video-context guided human mask inference and CNN based segmentation network learning iterate to mutually enhance each other until no further improvement gains. In the first step, each video is decomposed into supervoxels by the unsupervised video segmentation. The superpixels within the supervoxels are then classified as human or non-human by graph optimization with unary energies from the imperfect human detection results and the predicted confidence maps by the CNN trained in the previous iteration. In the second step, the video-context derived human masks are used as direct labels to train CNN. Extensive experiments on the challenging PASCAL VOC 2012 semantic segmentation benchmark demonstrate that the proposed framework has already achieved superior results than all previous weakly-supervised methods with object class or bounding box annotations. In addition, by augmenting with the annotated masks from PASCAL VOC 2012, our method reaches a new state-of-the-art performance on the human segmentation task.

Multi-fiber Networks for Video Recognition

In this paper, we aim to reduce the computational cost of spatio-temporal deep neural networks, making them run as fast as their 2D counterparts while preserving state-of-the-art accuracy on video recognition benchmarks. To this end, we present the novel Multi-Fiber architecture that slices a complex neural network into an ensemble of lightweight networks or fibers that run through the network. To facilitate information flow between fibers we further incorporate multiplexer modules and end up with an architecture that reduces the computational cost of 3D networks by an order of magnitude, while increasing recognition performance at the same time. Extensive experimental results show that our multi-fiber architecture significantly boosts the efficiency of existing convolution networks for both image and video recognition tasks, achieving state-of-the-art performance on UCF-101, HMDB-51 and Kinetics datasets. Our proposed model requires over 9\(\times \) and 13\(\times \) less computations than the I3D [1] and R(2+1)D [2] models, respectively, yet providing higher accuracy.

Collaborative Layer-Wise Discriminative Learning in Deep Neural Networks

Intermediate features at different layers of a deep neural network are known to be discriminative for visual patterns of different complexities. However, most existing works ignore such cross-layer heterogeneities when classifying samples of different complexities. For example, if a training sample has already been correctly classified at a specific layer with high confidence, we argue that it is unnecessary to enforce rest layers to classify this sample correctly and a better strategy is to encourage those layers to focus on other samples.

Training Group Orthogonal Neural Networks with Privileged Information

Learning rich and diverse representations is critical for the performance of deep convolutional neural networks (CNNs). In this paper, we consider how to use privileged information to promote inherent diversity of a single CNN model such that the model can learn better representations and offer stronger generalization ability. To this end, we propose a novel group orthogonal convolutional neural network (GoCNN) that learns untangled representations within each layer by exploiting provided privileged information and enhances representation diversity effectively. We take image classification as an example where image segmentation annotations are used as privileged information during the training process. Experiments on two benchmark datasets -- ImageNet and PASCAL VOC -- clearly demonstrate the strong generalization ability of our proposed GoCNN model. On the ImageNet dataset, GoCNN improves the performance of state-of-the-art ResNet-152 model by absolute value of 1.2% while only uses privileged information of 10% of the training images, confirming effectiveness of GoCNN on utilizing available privileged knowledge to train better CNNs.

3D-Assisted Coarse-to-Fine Extreme-Pose Facial Landmark Detection

We propose a novel 3D-assisted coarse-to-fine extreme-pose facial landmark detection system in this work. For a given face image, our system first refines the face bounding box with landmark locations inferred from a 3D face model generated by a Recurrent 3D Regressor at coarse level. Another R3R is then employed to fit a 3D face model onto the 2D face image cropped with the refined bounding box at fine-scale. 2D landmark locations inferred from the fitted 3D face are further adjusted with the popular 2D regression method, i.e. LBF. The 3D-assisted coarse-to-fine strategy and the 2D adjustment process explicitly ensure both the robustness to extreme face poses and bounding box disturbance and the accuracy towards pixel-level landmark displacement. Extensive experiments on the Menpo Challenge test sets demonstrate the superior performance of our system.

Multi-Human Parsing Machines

Human parsing is an important task in human-centric analysis. Despite the remarkable progress in single-human parsing, the more realistic case of multi-human parsing remains challenging in terms of the data and the model. Compared with the considerable number of available single-human parsing datasets, the datasets for multi-human parsing are very limited in number mainly due to the huge annotation effort required. Besides the data challenge to multi-human parsing, the persons in real-world scenarios are often entangled with each other due to close interaction and body occlusion, making it difficult to distinguish body parts from different person instances. In this paper we propose the Multi-Human Parsing Machines (MHPM) system, which contains an MHP Montage model and an MHP Solver, to address both challenges in multi-human parsing. Specifically, the MHP Montage model in MHPM generates realistic images with multiple persons together with the parsing labels. It intelligently composes single persons onto background scene images while maintaining the structural information between persons and the scene. The generated images can be used to train better multi-human parsing algorithms. On the other hand, the MHP Solver in MHPM solves the bottleneck of distinguishing multiple entangled persons with close interaction. It employs a Group-Individual Push and Pull (GIPP) loss function, which can effectively separate persons with close interaction. We experimentally show that the proposed MHPM can achieve state-of-the-art performance on the multi-human parsing benchmark and the person individualization benchmark, which distinguishes closely entangled person instances.