Hailin is a senior principal scientist at Adobe. He received his Master of Science and Doctor of Science degrees in electrical engineering from Washington University in Saint Louis in 2000 and 2003, respectively. Between fall 2003 and fall 2004, he was a postdoctoral researcher at the Computer Science Department, University of California at Los Angeles.
His current research interests include: deep learning, natural language processing, computer vision, video, image search, 3D reconstruction, structure and motion estimation, optical flow, stereo, and image-based modeling and rendering. His work can be found in several Adobe products including Photoshop, After Effects, Premiere Pro, Photoshop Lightroom, and Photoshop Elements.
For more information and a complete list of publications, please visit his personal page at https://sites.google.com/view/hailinjin or his Google Scholar profile at https://scholar.google.com/citations?user=DQOT0OMAAAAJ.
Interactive Boundary Prediction for Object Selection
Disentangling Structure and Aesthetics for Content-aware Image Completion
Spatial-Semantic Image Search by Visual Feature Synthesis
Composition-preserving Deep Photo Aesthetics Assessment
Rating Image Aesthetics Using Deep Learning
DeepFont: Identify Your Font from an Image
Selective Pooling Vector for Fine-Grained Recognition
Large-scale visual font recognition
GPU Asynchronous Stochastic Gradient Descent to Speed Up Neural Network Training
Automatic scene inference for 3D object compositing
Joint Subspace Stabilization for Stereoscopic Video
Large Displacement Optical Flow from Nearest Neighbor Fields
Plane-Based Content-Preserving Warps for Video Stabilization
Specular Reflection Separation using Dark Channel Priors
Causal Stereoscopic Photo Authoring
Content-Preserving Warps for Video Stabilization
Content-Preserving Warps for 3D Video Stabilization
Stereo Matching with Nonparametric Smoothness Priors in Feature Space
GroupSAC: Efficient Consensus in the Presence of Groupings
Light-Field Video Stabilization
a-three-point-minimal-solution-for-panoramic-stitching-with-lens-distortion
Stereoscopic Inpainting: Joint Color and Depth Completion from Stereo Images
3-D Reconstruction of Shaded Objects from Multiple Images under Known Illumination
Search Space Reduction for MRF Stereo
Mumford-Shah on the move: Region-based Segmentation on Deforming Manifolds with Application to 3-D Reconstruction of Shape and Appearance from Multi-view Images
Multi-view stereo reconstruction of dense shape and complex appearance
KALMANSAC: Robust filtering of concensus
Visual tracking in the presence of motion blur
Region-based segmentation on evolving surfaces with application to 3D reconstruction of shape and piecewise constant radiance
Shedding light on stereoscopic segmentation
Estimation of 3D surface shape and smooth radiance from 2D images: A level set approach
A semi-direct approach to structure from motion
Multi-view stereo beyond Lambert
Tales of shape and radiance in multi-view stereo
Variational methods for shape reconstruction in Computer Vision
A variational approach to shape from defocus
Structure from motion causally integrated over time
Variational multiframe stereo in the presence of specular reflections
A semi-direct approach to structure from motion
Real-time feature tracking and outlier rejection with changes in illumination
3-D motion and structure causally integrated over time: Implementation
Stereoscopic shading: Integrating multi-frame shape cues in a variational framework