Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks56 from public data
- A view-free image stitching network based on global homography155
A cascaded view-free image stitching network based on a global homography, which can achieve almost 100% elimination of artifacts in overlapping areas at the cost of acceptable slight distortions in non-overlapping areas, compared with traditional methods.
- Parallax-Tolerant Unsupervised Deep Image Stitching134
This work proposes a parallax-tolerant unsupervised deep image stitching technique that is parallax-tolerant and free from laborious designs of complicated geometric features for specific scenes, both quantitatively and qualitatively.
- 117
- Depth-Aware Multi-Grid Deep Homography Estimation With Contextual Correlation108
A contextual correlation layer (CCL) is designed that can efficiently capture the long-range correlation within feature maps and can be flexibly used in a learning framework to predict multi-grid homography from global to local.
- 79
- 68
- 68
- Progressively Complementary Network for Fisheye Image Rectification Using Appearance Flow65
This paper embeds a correction layer in skipconnection and leverage the appearance flows in different layers to pre-correct the image features and proposes a parallel complementary structure that effectively reduces the burden of the decoder by separating content reconstruction and structure correction.
- 46
- Pseudo-LiDAR Point Cloud Interpolation Based on 3D Motion Representation and Spatial Supervision40
This work designs a novel reconstruction loss function with the chamfer distance to supervise the generation of Pseudo-LiDAR point clouds in 3D space and introduces a multi-modal deep aggregation module to facilitate the efficient fusion of texture and depth features.
- A Deep Ordinal Distortion Estimation Approach for Distortion Rectification40
This work proposes a novel distortion rectification approach that can obtain more accurate parameters with higher efficiency, and is the first to unify the heterogeneous distortion parameters into a learning-friendly intermediate representation through ordinal distortion, bridging the gap between image feature and distort rectification.
- Unsupervised fisheye image correction through bidirectional loss with geometric prior28
The proposed unsupervised fisheye correction network is more flexible and shows better practical applications for distortion rectification, and outperforms state-of-the-art methods on the correction performance without any labeled distortion parameters.
- Complementary Bi-directional Feature Compression for Indoor 360° Semantic Segmentation with Self-distillation24
This paper combines the two different representations and proposes a novel 360° semantic segmentation solution from a complementary perspective, which outperforms the state-of-the-art solutions on quantitative evaluations while displaying the best performance on visual appearance.
- PLIN: A Network for Pseudo-LiDAR Point Cloud Interpolation24
A novel Pseudo-LiDAR interpolation network (PLIN) to increase the frequency of LiDAR sensor data and design a coarse interpolation stage guided by consecutive sparse depth maps and motion relationship that can progressively perceive multi-modal information and generate accurate intermediate point clouds.
- 23
- 22
- Semi-Supervised Coupled Thin-Plate Spline Model for Rotation Correction and Beyond21
The coupled thin-plate spline model (CoupledTPS), which iteratively couples multiple TPS with limited control points into a more flexible and powerful transformation, and a semi-supervised learning scheme to improve warping quality by exploiting unlabeled data is developed.
- Disentangling Orthogonal Planes for Indoor Panoramic Room Layout Estimation with Cross-Scale Distortion Awareness21
This work proposes to disentangle this 1D representation by pre-segmenting orthogonal (vertical and horizontal) planes from a complex scene, explicitly capturing the geometric cues for indoor layout estimation.
- MOWA: Multiple-in-One Image Warping Model19
This work proposes a Multiple-in-One image WArping model (named MOWA), the first work that solves multiple practical warping tasks in one single model and introduces a lightweight point-based classifier that predicts the task type and serves as prompts to modulate the feature maps for more accurate estimation.
- Cylin-Painting: Seamless 360° Panoramic Image Outpainting and Beyond19
A deep analysis of the differences between inpainting and outpainting is provided, which essentially depends on how the source pixels contribute to the unknown regions under different spatial arrangements, and a learnable positional embedding strategy to incorporate the missing component of positional encoding into the cylinder convolution, which significantly improves the panoramic results.
- Distortion-Tolerant Monocular Depth Estimation on Omnidirectional Images Using Dual-Cubemap18
A distortion-tolerant omnidirectional depth estimation algorithm using a dual-cubemap model to reduce the distortion at the cost of boundary discontinuity on omniddirectional depths and a boundary revision module to smooth the discontinuous boundaries.
- Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration17
This paper shows that it is possible to perform domain adaptation via the noise space using diffusion models and derives a meaningful diffusion loss that guides the restoration model in progressively aligning both restored synthetic and real-world outputs with a target clean distribution.
- 17
- Deep Rotation Correction Without Angle Prior16
This paper proposes a new and practical task, named Rotation Correction, to automatically correct the tilt with high content fidelity in the condition that the rotated angle is unknown, and demonstrates that the algorithm can outperform other state-of-the-art solutions requiring this prior.
- 15
- Unsupervised OmniMVS: Efficient Omnidirectional Depth Inference via Establishing Pseudo-Stereo Supervision14
Experiments exhibit that the performance of the unsupervised omnidirectional MVS framework is competitive to that of the state-of-the-art (SoTA) supervised methods with better generalization in real-world data.
- Neural Contourlet Network for Monocular 360° Depth Estimation14
A new perspective that constructs an interpretable and sparse representation for a 360° image by utilizing the contourlet transform to capture an explicit geometric cue in the spectral domain and integrate it with an implicit cues in the spatial domain is provided.
- 14
- FishFormer: Annulus Slicing-based Transformer for Fisheye Rectification13
FishFormer is introduced that processes the fisheye image as a sequence to enhance global and local perception and a novel layer attention mechanism is introduced to enhance the local perception and feature interaction.
- S-OmniMVS: Incorporating Sphere Geometry into Omnidirectional Stereo Matching11
This paper revisits omnidirectional MVS by incorporating three sphere geometry priors: spherical projection, spherical continuity, and spherical position, and develops a segmented sampling strategy that combines linear and exponential spaces to create S-OmniMVS, along with three sphere priors.
- 11
- 10
- 9
- 9
- Pseudo-LiDAR point cloud magnification9
This paper presents a novel pseudo-LiDAR point cloud magnification algorithm, aiming to extrapolate the narrow baseline and further bridge the gap between LiDAR and camera, and proposes a region-aware restoration approach to obtain more realistic synthesized results.
- 8
- 360 Layout Estimation via Orthogonal Planes Disentanglement and Multi-View Geometric Consistency Perception8
An orthogonal plane disentanglement network (termed DOPNet) is proposed to distinguish ambiguous semantics and an unsupervised adaptation technique tailored for horizon-depth and ratio representations is presented, which outperforms other SoTA models on both monocular layout estimation and multi-view layout estimation tasks.
- 8
- Revisiting 360 Depth Estimation With PanoGabor: A New Fusion Perspective7
This work introduces Gabor filters that analyze texture in the frequency domain, extending the receptive fields and enhancing depth cues, and designs a channel-wise and spatial-wise unidirectional fusion module (CS-UFM) that integrates the proposed PanoGabor filters to unify other representations into the ERP format, delivering effective and distortion-aware features.
- Robust Image Stitching With Optimal Plane5
It is proposed to incorporate the universal prior of content perception into the image stitching model by a dual-branch architecture, which separately captures coarse and fine features and integrates them to achieve highly generalizable performance across diverse unseen real-world scenes.
- 5
- 4
- 3
- 3
- 3
- 2
- Towards Reliable Image Outpainting: Learning Structure-Aware Multimodal Fusion with Depth Guidance2
A reliable image outpainting task, introducing the sparse depth from LiDARs to extrapolate authentic RGB scenes and specially design an additional constraint strategy consisting of Cross-modal Loss and Edge Loss to enhance ambiguous contours and expedite reliable content generation.
- 1
- 1
- –
- –
- –
- Semi-Supervised 360 Layout Estimation With Panoramic Collaborative Perturbations–
This work proposes a novel semi-supervised method named SemiLayout360, which incorporates the priors of the panoramic layout and distortion through collaborative perturbations, and introduces the panoramic distortion prior to strengthen distortion awareness.
- –
- Knowledge Visualization: A Benchmark and Method for Knowledge-Intensive Text-to-Image Generation–
KVBench is introduced, a curriculum-grounded benchmark for evaluating knowledge-intensive T2I generation and KE-Check, a two-stage framework that improves scientific fidelity via Knowledge Elaboration for structured prompt enrichment and Checklist-Guided Refinement for explicit constraint enforcement through violation identification and constraint-guided editing.
- Beyond Wide-Angle Images: Structure-to-Detail Video Portrait Correction via Unsupervised Spatiotemporal Adaptation–
A structure-to-detail portrait correction model named ImagePC that integrates the long-range awareness of the transformer and multi-step denoising of diffusion models into a unified framework, achieving global structural robustness and local detail refinement and contributing to high-fidelity wide-angle videos with stable and natural portraits.
Publication data from OpenAlex; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-10. 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. You will be asked to sign in.