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
- Xiaoou TangSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works72 from public data
- BiFormer: Vision Transformer with Bi-Level Routing Attention1,180
This work proposes a novel dynamic sparse attention via bi-level routing to enable a more flexible allocation of computations with content awareness and presents a new general vision transformer, named BiFormer, which enjoys both good performance and high computational efficiency, especially in dense prediction tasks.
- ViM: Out-Of-Distribution with Virtual-logit Matching594
A novel OOD scoring method named Virtual-logit Matching (ViM), which combines the class-agnostic score from feature space and the In-Distribution (ID) class-dependent logits, which is 4% ahead of the best baseline.
- Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation577
This work revisits the weak-to-strong consistency framework, popularized by FixMatch from semi-supervised classification, and presents a dual-stream perturbation technique, enabling two strong views to be simultaneously guided by a common weak view.
- 404
- Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognition342
A novel compact motion representation for video action recognition, named Optical Flow guided Feature (OFF), which enables the network to distill temporal information through a fast and robust approach and is complementary to other motion modalities such as optical flow.
- Fourier Contour Embedding for Arbitrary-Shaped Text Detection310
Extensive experiments demonstrate that FCE is accurate and robust to fit contours of scene texts even with highly-curved shapes, and also validate the effectiveness and the good generalization of FCENet for arbitrary-shaped text detection.
- RobustScanner: Dynamically Enhancing Positional Clues for Robust Text Recognition200
Theoretically, the proposed method, dubbed \emph{RobustScanner}, decodes individual characters with dynamic ratio between context and positional clues, and utilizes more positional ones when the decoding sequences with scarce context, and thus is robust and practical.
- Graph Degree Linkage: Agglomerative Clustering on a Directed Graph179
This paper proposes a simple but effective graph-based agglomerative algorithm, for clustering high-dimensional data, and defines the affinity measure of clusters via the product of average indegree and average outdegree.
- Semantically Coherent Out-of-Distribution Detection161
This work re-designs the benchmarks and proposes the semantically coherent out-of-distribution detection (SC-OOD), and designs a concise framework featured by unsupervised dual grouping (UDG) for the joint modeling of ID and OOD data.
- Scale-Equalizing Pyramid Convolution for Object Detection132
A scale-equalizing pyramid Convolution (SEPC) that aligns the shared pyramid convolution kernel only at high-level feature maps is built, which brings significant performance improvement in state-of-the-art one-stage object detectors and a light version of SEPC also has ~3.5AP gain with only around 7% inference time increase.
- Agglomerative clustering via maximum incremental path integral128
A novel graph-structural agglomerative clustering algorithm, where the graph encodes local structures of data and the affinity between two clusters is defined as Incremental Path Integral, which can be computed in a closed-form exact solution, with linear time complexity with respect to the maximum size of clusters.
- Consistent-Teacher: Towards Reducing Inconsistent Pseudo-Targets in Semi-Supervised Object Detection123
A adaptive anchor assignment (ASA) substitutes the static IoU-based strategy, which enables the student network to be resistant to noisy pseudo-bounding boxes, and a Gaussian Mixture Model dynamically revises the score threshold of pseudo-bboxes, which stabilizes the number of ground truths at an early stage.
- Vision Transformer with Progressive Sampling121
An iterative and progressive sampling strategy to locate discriminative regions and when combined with the Vision Transformer, the obtained PS-ViT network can adaptively learn where to look.
- Enabling Multilevel Trust in Privacy Preserving Data Mining117
The solution prevents data miners from jointly reconstructing the original data more accurately than the best effort using any individual copy in the collection, and allows a data owner to generate perturbed copies of its data for arbitrary trust levels on-demand, which offers data owners maximum flexibility.
- 115
- Pseudo-mask Matters in Weakly-supervised Semantic Segmentation113
The following designs are proposed to push the performance of weakly supervised semantic segmentation methods to new state-of-art: coefficient of Variation Smoothing to smooth the CAMs adaptively, andportional Pseudo-mask Generation to project the expanded CAMs to pseudo-mask based on a new metric indicating the importance of each class on each location, instead of the scores trained from binary classifiers.
- Lighting and Pose Robust Face Sketch Synthesis109
A robust algorithm for synthesizing a face sketch from a face photo taken under a different lighting condition and in a different pose than the training set is proposed using a multiscale Markov Random Field (MRF) model.
- Fashion Retrieval via Graph Reasoning Networks on a Similarity Pyramid107
A novel Graph Reasoning Network (GRNet) on a Similarity Pyramid, which learns similarities between a query and a gallery cloth by using both global and local representations in multiple scales, and achieves considerable improvements on all empirical settings.
- Uncertainty Estimation via Response Scaling for Pseudo-Mask Noise Mitigation in Weakly-Supervised Semantic Segmentation102
This paper simulates noisy variations of response by scaling the prediction map in multiple times for uncertainty estimation, and the uncertainty is then used to weight the segmentation loss to mitigate noisy supervision signals.
- MMOCR95
MMOCR is an open-source toolbox which provides a comprehensive pipeline for text detection and recognition, as well as their downstream tasks such as named entity recognition and key information extraction.
- Object Instance Mining for Weakly Supervised Object Detection93
An end-to-end object instance mining (OIM) framework for weakly supervised object detection that attempts to detect all possible object instances existing in each image by introducing information propagation on the spatial and appearance graphs, without any additional annotations is introduced.
- 88
- Context-Aware RCNN: A Baseline for Action Detection in Videos87
This work empirically finds the recognition accuracy is highly correlated with the bounding box size of an actor, and thus higher resolution of actors contributes to better performance, and develops a surpringly effective baseline (Context-Aware RCNN) that achieves new state-of-the-art results on two challenging action detection benchmarks of AVA and JHMDB.
- Panoptic Video Scene Graph Generation73
A high-quality PVSG dataset is contributed, which consists of 400 videos (289 third-person + 111 egocentric videos) with totally 150K frames labeled with panoptic segmentation masks as well as fine, temporal scene graphs.
- Text4Seg: Reimagining Image Segmentation as Text Generation70
Text4Seg is introduced, a novel text-as-mask paradigm that casts image segmentation as a text generation problem, eliminating the need for additional decoders and significantly simplifying the segmentation process.
- 70
- Fast Video Shot Transition Localization with Deep Structured Models60
A structured network aiming to detect these two shot transitions using targeted models separately using one TITAN GPU and Experiments on public TRECVID07 and RAI databases show that the method outperforms the state-of-the-art methods.
- GeoGround: A Unified Large Vision-Language Model for Remote Sensing Visual Grounding59
GeoGround is a novel framework that unifies support for HBB, OBB, and mask RS visual grounding tasks, allowing flexible output selection through the Text-Mask technique, and defines prompt-assisted and geometry-guided learning to enhance consistency across different signals.
- Extractive Video Summarizer with Memory Augmented Neural Networks59
A memory augmented extractive video summarizer, which utilizes an external memory to record visual information of the whole video with high capacity and demonstrates that the global attention modeling has two advantages: good transferring ability across datasets and high robustness to noisy videos.
- Learning PDEs for Image Restoration via Optimal Control57
The design of PDEs could be made easier by borrowing the learning strategy from machine learning, and the effectiveness of the learning-based PDE (L-PDE) framework for image restoration is demonstrated.
- 51
- 45
- 43
- Weakly Supervised Video Salient Object Detection via Point Supervision41
A hybrid token attention module is proposed, which mixes optical flow and image information from orthogonal directions, adaptively highlighting critical optical flow information (channel dimension) and critical token information (spatial dimension) to infer saliency maps with temporal information.
- Recovery of Future Data via Convolution Nuclear Norm Minimization33
This work converts TSF into a more inclusive problem called tensor completion with arbitrary sampling (TCAS), and proves that CNNM succeeds in solving TCAS as long as a sampling condition—which depends on the convolution rank of the target tensor—is obeyed.
- Geometry Normalization Networks for Accurate Scene Text Detection33
A novel Geometry Normalization Module with multiple branches to normalize each text instance to one desired canonical geometry range through at least one branch, and proposes a geometry-aware training scheme to effectively train the GNNets by sampling and augmenting text instances from a uniform geometry variance distribution.
- 32
- Tensor linear Laplacian discrimination (TLLD) for feature extraction32
Experimental results on face recognition, texture classification and handwritten digit recognition show that the tensor linear Laplacian discrimination algorithm TLLD is effective in extracting discriminative features.
- 29
- 28
- 28
- 26
- 26
- 25
- 25
- 24
- 19
- 18
- Webly Supervised Image Classification with Self-contained Confidence16
Inspired by the ability of Deep Neural Networks (DNNs) in confidence prediction, Self-Contained Confidence (SCC) is introduced by adapting model uncertainty for WSL setting, and used to sample-wisely balance the balance between web label and supervised loss on sample level.
- 15
- Webly Supervised Image Classification with Metadata: Automatic Noisy Label Correction via Visual-Semantic Graph15
VSGraph-LC starts from anchor selection referring to the semantic similarity between metadata and correct label concepts, and then propagates correct labels from anchors on a visual graph using graph neural network (GNN).
- 14
- 12
- 10
- 10
- 8
- 7
- 7
- 6
- 5
- 5
- 4
- 3
- 3
- 3
- 3
- 2
- 2
- 2
- 1
- 1
- –
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 2026-10-13. 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.