Tianli Liao

I am a lecturer in the College of Information Science and Engineering, Henan University of Technology. I received my Ph.D. degree in Applied Mathematics in 2019 from the Center for Combinatorics, Nankai University, supervised by William Y.C. Chen. My current research interests mainly lie in the fields of deep learning, computer vision, and image processing, especially image/video stitching, image manipulation, and neural rendering.

Email  /  Scholar  /  Github

Local Patch Alignment Module (LPAM) is proposed to improve the performance of seam-cutting methods for large parallax image stitching.

Using (ground truth or estimated) depth maps to construct epipolar geometry between large parallax images to achieve better image alignment and warping results.

Leveraging SAM (Segment Anything Model) into multiple image warping algorithm for large parallax image stitching.

We propose two single-perspective warps for natural image stitching. The first one is a parametric warp, the second one is a mesh-based warp.

We proposed a novel iterative seam estimation method where the iteration procedure is guided by our quality evaluation for the pixels along the seam.

A novel homography geometric matching architecture with homography prior is proposed.

A perception-based energy function is designed for seam-cutting method, in which the nonlinearity and the nonuniformity of human perception is considered.

• Journal Reviewer: TVCG, TIP, TCSVT

awesome website template