Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Suggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works207 from public data
- Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection1,828
The proposed memory-augmented autoencoder called MemAE is free of assumptions on the data type and thus general to be applied to different tasks and proves the excellent generalization and high effectiveness of the proposed MemAE.
- Graph-Structured Representations for Visual Question Answering496
This paper proposes to build graphs over the scene objects and over the question words, and describes a deep neural network that exploits the structure in these representations, and achieves significant improvements over the state-of-the-art.
- 493
- What Value Do Explicit High Level Concepts Have in Vision to Language Problems?482
A method of incorporating high-level concepts into the successful CNN-RNN approach is proposed, and it is shown that it achieves a significant improvement on the state-of-the-art in both image captioning and visual question answering.
- From Motion Blur to Motion Flow: A Deep Learning Solution for Removing Heterogeneous Motion Blur464
This work directly estimates the motion flow from the blurred image through a fully-convolutional deep neural network (FCN) and recovers the unblurred image from the estimated motion flow and is the first universal end-to-end mapping from the blur image to the dense motion flow.
- Deep learning features at scale for visual place recognition366
This paper trains, at large scale, two CNN architectures for the specific place recognition task and employs a multi-scale feature encoding method to generate condition- and viewpoint-invariant features.
- Adversarial PoseNet: A Structure-Aware Convolutional Network for Human Pose Estimation360
This work proposes a novel structure-aware convolutional network to implicitly take priors about the structure of human bodies into account during training of the deep network and significantly outperforms the state-of-the-art methods and almost always generates plausible human pose predictions.
- Self-supervised Learning of Adversarial Example: Towards Good Generalizations for Deepfake Detection346
This work addresses the generalizable deepfake detection from a simple principle: a generalizable representation should be sensitive to diverse types of forgeries and synthesize augmented forgeries with a pool of forgery configurations and strengthen the “sensitivity” to the forgeries by enforcing the model to predict the forgery configuration.
- ZegCLIP: Towards Adapting CLIP for Zero-shot Semantic Segmentation325
A simpler-and-efficient one-stage solution that directly extends CLIP's zero-shot prediction capability from image to pixel level and demonstrates superior performance, outperforming the state-of-the-art methods by a large margin under both “inductive” and “transductive” zero- shot settings.
- Towards Effective Low-Bitwidth Convolutional Neural Networks266
This paper tackles the problem of training a deep convolutional neural network with both low-precision weights and low-bitwidth activations by proposing a two-stage optimization strategy to progressively find good local minima and adopting a novel learning scheme to jointly train a full- Precision model alongside the low-Precision one.
- METransformer: Radiology Report Generation by Transformer with Multiple Learnable Expert Tokens232
The proposed METransformer, a method to realize a “multi-expert joint diagnosis” mechanism to upgrade the existing “single expert” framework commonly seen in the current literature, enjoys the merits of an ensemble-based approach but through a manner that is computationally more efficient and supports more sophisticated interactions among experts.
- The treasure beneath convolutional layers: Cross-convolutional-layer pooling for image classification201
This paper advocates that if used appropriately, convolutional layer activations constitute a powerful image representation by adopting a new technique proposed in this paper called cross-convolutional-layer pooling, and can achieve comparable or in some cases significantly better performance than existing fully-connected layer based image representations.
- Less is More: Zero-Shot Learning from Online Textual Documents with Noise Suppression194
An l2,1-norm based objective function is proposed which can simultaneously suppress the noisy signal in the text and learn a function to match the text document and visual features and develops an optimization algorithm to efficiently solve the resulting problem.
- Seeing Deeply and Bidirectionally: A Deep Learning Approach for Single Image Reflection Removal191
This work argues that, to remove reflection truly well, it should estimate the reflection and utilize it to estimate the background image, and proposes a cascade deep neural network, which estimates both the Background image and the reflection.
- 189
- Feature Encoding With Autoencoders for Weakly Supervised Anomaly Detection178
This article proposes a novel strategy to transform the input data into a more meaningful representation that could be used for anomaly detection and leverages an autoencoder to encode the data and utilizes three factors, hidden representation, reconstruction residual vector, and reconstruction error, as the new representation for the inputData.
- Towards Context-Aware Interaction Recognition for Visual Relationship Detection168
This paper proposes an alternative, context-aware interaction recognition framework, and explicitly construct an interaction classifier which combines the context, and the interaction, to avoid the limitation of both approaches.
- Structured Binary Neural Networks for Accurate Image Classification and Semantic Segmentation166
This paper proposes a ``network decomposition'' strategy, named Group-Net, in which each full-precision group can be effectively reconstructed by aggregating a set of homogeneous binary branches, and shows strong generalization to other tasks.
- Piecewise Classifier Mappings: Learning Fine-Grained Learners for Novel Categories With Few Examples163
An end-to-end trainable deep network inspired by the state-of-the-art fine-grained recognition model and is tailored for the FSFG task is proposed, which generates the decision boundary via learning a set of more attainable sub-classifiers in a more parameter-economic way.
- Temporal Pyramid Pooling-Based Convolutional Neural Network for Action Recognition157
A novel network structure, which allows an arbitrary number of frames as the network input, is proposed and can be learned on a small target data set because it can leverage the off-the-shelf image-level CNN for model parameter initialization.
- Global and Local Texture Randomization for Synthetic-to-Real Semantic Segmentation128
Two simple yet effective texture randomization mechanisms, GTR and LTR, are proposed to randomize the texture of source images into diverse unreal texture styles to alleviate the reliance of the network on texture while promoting the learning of the domain-invariant cues.
- Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularization122
This work proposes a novel approach to regularize the distribution of within-class features to ease label propagation difficulty and creates a semi-supervised semantic segmentation algorithm that demonstrates superior performance over the state-of-the-art methods from extensive experimental evaluation on both Pascal VOC and Cityscapes benchmarks.
- 119
- Towards using count-level weak supervision for crowd counting114
This paper studies the problem of weakly-supervised crowd counting, and devise a simple-yet-effective training strategy, namely Multiple Auxiliary Tasks Training (MATT), to construct regularizes for restricting the freedom of the generated density maps.
- Mid-level Deep Pattern Mining∗114
This work finds that for an image patch, activation extracted from the first fully-connected layer of a CNN have two appealing properties which enable its seamless integration with pattern mining, and applies this approach to scene and object classification tasks, demonstrating that it outperforms all previous works on mid-level visual element discovery by a sizeable margin.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.