Ce Liu

Bio: Ce Liu is an academic researcher from Google. The author has contributed to research in topics: Computer science & Optical flow. The author has an hindex of 47, co-authored 110 publications receiving 13484 citations. Previous affiliations of Ce Liu include Adobe Systems & Massachusetts Institute of Technology.

SIFT Flow: Dense Correspondence across Scenes and Its Applications

Abstract: While image alignment has been studied in different areas of computer vision for decades, aligning images depicting different scenes remains a challenging problem. Analogous to optical flow, where an image is aligned to its temporally adjacent frame, we propose SIFT flow, a method to align an image to its nearest neighbors in a large image corpus containing a variety of scenes. The SIFT flow algorithm consists of matching densely sampled, pixelwise SIFT features between two images while preserving spatial discontinuities. The SIFT features allow robust matching across different scene/object appearances, whereas the discontinuity-preserving spatial model allows matching of objects located at different parts of the scene. Experiments show that the proposed approach robustly aligns complex scene pairs containing significant spatial differences. Based on SIFT flow, we propose an alignment-based large database framework for image analysis and synthesis, where image information is transferred from the nearest neighbors to a query image according to the dense scene correspondence. This framework is demonstrated through concrete applications such as motion field prediction from a single image, motion synthesis via object transfer, satellite image registration, and face recognition.

Supervised Contrastive Learning.

Abstract: Contrastive learning applied to self-supervised representation learning has seen a resurgence in recent years, leading to state of the art performance in the unsupervised training of deep image models. Modern batch contrastive approaches subsume or significantly outperform traditional contrastive losses such as triplet, max-margin and the N-pairs loss. In this work, we extend the self-supervised batch contrastive approach to the fully-supervised setting, allowing us to effectively leverage label information. Clusters of points belonging to the same class are pulled together in embedding space, while simultaneously pushing apart clusters of samples from different classes. We analyze two possible versions of the supervised contrastive (SupCon) loss, identifying the best-performing formulation of the loss. On ResNet-200, we achieve top-1 accuracy of 81.4% on the ImageNet dataset, which is 0.8% above the best number reported for this architecture. We show consistent outperformance over cross-entropy on other datasets and two ResNet variants. The loss shows benefits for robustness to natural corruptions and is more stable to hyperparameter settings such as optimizers and data augmentations. Our loss function is simple to implement, and reference TensorFlow code is released at this https URL.

Beyond pixels: exploring new representations and applications for motion analysis

Abstract: The focus of motion analysis has been on estimating a flow vector for every pixel by matching intensities. In my thesis, I will explore motion representations beyond the pixel level and new applications to which these representations lead. I first focus on analyzing motion from video sequences. Traditional motion analysis suffers from the inappropriate modeling of the grouping relationship of pixels and from a lack of ground-truth data. Using layers as the interface for humans to interact with videos, we build a human-assisted motion annotation system to obtain ground-truth motion, missing in the literature, for natural video sequences. Furthermore, we show that with the layer representation, we can detect and magnify small motions to make them visible to human eyes. Then we move to a contour presentation to analyze the motion for textureless objects under occlusion. We demonstrate that simultaneous boundary grouping and motion analysis can solve challenging data, where the traditional pixel-wise motion analysis fails. In the second part of my thesis, I will show the benefits of matching local image structures instead of intensity values. We propose SIFT flow that establishes dense, semantically meaningful correspondence between two images across scenes by matching pixel-wise SIFT features. Using SIFT flow, we develop a new framework for image parsing by transferring the metadata information, such as annotation, motion and depth, from the images in a large database to an unknown query image. We demonstrate this framework using new applications such as predicting motion from a single image and motion synthesis via object transfer. Based on SIFT flow, we introduce a nonparametric scene parsing system using label transfer, with very promising experimental results suggesting that our system outperforms state-of-the-art techniques based on training classifiers. (Copies available exclusively from MIT Libraries, Rm. 14-0551, Cambridge, MA 02139-4307. Ph. 617-253-5668; Fax 617-253-1690.)

Deep Convolutional Neural Network for Image Deconvolution

Abstract: Many fundamental image-related problems involve deconvolution operators. Real blur degradation seldom complies with an ideal linear convolution model due to camera noise, saturation, image compression, to name a few. Instead of perfectly modeling outliers, which is rather challenging from a generative model perspective, we develop a deep convolutional neural network to capture the characteristics of degradation. We note directly applying existing deep neural networks does not produce reasonable results. Our solution is to establish the connection between traditional optimization-based schemes and a neural network architecture where a novel, separable structure is introduced as a reliable support for robust deconvolution against artifacts. Our network contains two submodules, both trained in a supervised manner with proper initialization. They yield decent performance on non-blind image deconvolution compared to previous generative-model based methods.

Real-time texture synthesis by patch-based sampling

Abstract: We present an algorithm for synthesizing textures from an input sample. This patch-based sampling algorithm is fast and it makes high-quality texture synthesis a real-time process. For generating textures of the same size and comparable quality, patch-based sampling is orders of magnitude faster than existing algorithms. The patch-based sampling algorithm works well for a wide variety of textures ranging from regular to stochastic. By sampling patches according to a nonparametric estimation of the local conditional MRF density function, we avoid mismatching features across patch boundaries. We also experimented with documented cases for which pixel-based nonparametric sampling algorithms cease to be effective but our algorithm continues to work well.

ImageNet Large Scale Visual Recognition Challenge

Abstract: The ImageNet Large Scale Visual Recognition Challenge is a benchmark in object category classification and detection on hundreds of object categories and millions of images. The challenge has been run annually from 2010 to present, attracting participation from more than fifty institutions. This paper describes the creation of this benchmark dataset and the advances in object recognition that have been possible as a result. We discuss the challenges of collecting large-scale ground truth annotation, highlight key breakthroughs in categorical object recognition, provide a detailed analysis of the current state of the field of large-scale image classification and object detection, and compare the state-of-the-art computer vision accuracy with human accuracy. We conclude with lessons learned in the 5 years of the challenge, and propose future directions and improvements.

Fully convolutional networks for semantic segmentation

Abstract: Convolutional networks are powerful visual models that yield hierarchies of features. We show that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, exceed the state-of-the-art in semantic segmentation. Our key insight is to build “fully convolutional” networks that take input of arbitrary size and produce correspondingly-sized output with efficient inference and learning. We define and detail the space of fully convolutional networks, explain their application to spatially dense prediction tasks, and draw connections to prior models. We adapt contemporary classification networks (AlexNet [20], the VGG net [31], and GoogLeNet [32]) into fully convolutional networks and transfer their learned representations by fine-tuning [3] to the segmentation task. We then define a skip architecture that combines semantic information from a deep, coarse layer with appearance information from a shallow, fine layer to produce accurate and detailed segmentations. Our fully convolutional network achieves state-of-the-art segmentation of PASCAL VOC (20% relative improvement to 62.2% mean IU on 2012), NYUDv2, and SIFT Flow, while inference takes less than one fifth of a second for a typical image.

SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation

Abstract: We present a novel and practical deep fully convolutional neural network architecture for semantic pixel-wise segmentation termed SegNet. This core trainable segmentation engine consists of an encoder network, a corresponding decoder network followed by a pixel-wise classification layer. The architecture of the encoder network is topologically identical to the 13 convolutional layers in the VGG16 network [1] . The role of the decoder network is to map the low resolution encoder feature maps to full input resolution feature maps for pixel-wise classification. The novelty of SegNet lies is in the manner in which the decoder upsamples its lower resolution input feature map(s). Specifically, the decoder uses pooling indices computed in the max-pooling step of the corresponding encoder to perform non-linear upsampling. This eliminates the need for learning to upsample. The upsampled maps are sparse and are then convolved with trainable filters to produce dense feature maps. We compare our proposed architecture with the widely adopted FCN [2] and also with the well known DeepLab-LargeFOV [3] , DeconvNet [4] architectures. This comparison reveals the memory versus accuracy trade-off involved in achieving good segmentation performance. SegNet was primarily motivated by scene understanding applications. Hence, it is designed to be efficient both in terms of memory and computational time during inference. It is also significantly smaller in the number of trainable parameters than other competing architectures and can be trained end-to-end using stochastic gradient descent. We also performed a controlled benchmark of SegNet and other architectures on both road scenes and SUN RGB-D indoor scene segmentation tasks. These quantitative assessments show that SegNet provides good performance with competitive inference time and most efficient inference memory-wise as compared to other architectures. We also provide a Caffe implementation of SegNet and a web demo at http://mi.eng.cam.ac.uk/projects/segnet/ .

Pyramid Scene Parsing Network

Abstract: Scene parsing is challenging for unrestricted open vocabulary and diverse scenes. In this paper, we exploit the capability of global context information by different-region-based context aggregation through our pyramid pooling module together with the proposed pyramid scene parsing network (PSPNet). Our global prior representation is effective to produce good quality results on the scene parsing task, while PSPNet provides a superior framework for pixel-level prediction. The proposed approach achieves state-of-the-art performance on various datasets. It came first in ImageNet scene parsing challenge 2016, PASCAL VOC 2012 benchmark and Cityscapes benchmark. A single PSPNet yields the new record of mIoU accuracy 85.4% on PASCAL VOC 2012 and accuracy 80.2% on Cityscapes.

Fully Convolutional Networks for Semantic Segmentation

Abstract: Convolutional networks are powerful visual models that yield hierarchies of features. We show that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, exceed the state-of-the-art in semantic segmentation. Our key insight is to build "fully convolutional" networks that take input of arbitrary size and produce correspondingly-sized output with efficient inference and learning. We define and detail the space of fully convolutional networks, explain their application to spatially dense prediction tasks, and draw connections to prior models. We adapt contemporary classification networks (AlexNet, the VGG net, and GoogLeNet) into fully convolutional networks and transfer their learned representations by fine-tuning to the segmentation task. We then define a novel architecture that combines semantic information from a deep, coarse layer with appearance information from a shallow, fine layer to produce accurate and detailed segmentations. Our fully convolutional network achieves state-of-the-art segmentation of PASCAL VOC (20% relative improvement to 62.2% mean IU on 2012), NYUDv2, and SIFT Flow, while inference takes one third of a second for a typical image.