Yanlin Qian

I am a color&luma ISP research engineer, with my enthusiasm on making realtime low-ppa Computational photography algorithms and its product implementation. I work for DJI/HasselBlad now.

I am an ex-Huawei&Honor&MicroBT engineer. I work on a bunch but still quite narrow ISP problems, for example auto-white balance (awb), illumination reflentance spectra estimation, chromatic adaption, blc lsc ccm ae and so on. I graduated from Phd in Tampere University in December 2020, where I was supervised by Prof. Joni-Kristian Kämäräinen and Prof. Jiri Matas.

Email  /  Google Scholar  /  Linkedin

For the task of edge detection in a noisy setting, we propose an unadorned method, called PPCE, modifying the classic Canny Edge by using a Pull-Push non-local means (PP-NLM) denoising framework.

Current 2D-to-3D conversion methods achieve geometric accuracy but are artistically deficient, failing to replicate the immersive and emotionally resonant experience of professional 3D cinema. This is because "geometric reconstruction" paradigms mistake deliberate artistic intent—such as strategic zero-plane shifts for "pop-out" effects and local depth sculpting—for data "noise" or ambiguity. This paper argues for a new paradigm: \textbf{Artistic Disparity Synthesis}, shifting the goal from physically accurate disparity estimation to artistically coherent disparity synthesis. We propose Art3D, a preliminary framework exploring this paradigm.

We propose a novel research question: how to employ events to enhance and adaptively adjust the brightness of images captured under broad lighting conditions?

we present a new event-RAW paired dataset, collected with a novel but still confidential sensor that records pixel-level aligned events and RAW images. This dataset includes 3373 RAW images with 2248×3264 resolution and their corresponding events, spanning 24 scenes with 3 exposure modes and 3 lenses. Second, we propose a conventional ISP pipeline to generate good RGB frames as reference.

We propose Integral Fast Fourier Color Constancy (IFFCC), an extension of FFCC tailored for multi-illuminant scenes.

we propose a novel denoising diffusion model based on shortest-path modeling that optimizesresidual propagation to enhance both denoising efficiency and quality. Drawing on Denoising Diffusion Implicit Models (DDIM) and insights from graph theory, our model, termed the Shortest Path Diffusion Model (ShortDF), treats the denoising process as a shortest-path problem aimed at minimizing reconstruction error. By optimizing the initial residuals, we improve the efficiency of the reverse diffusion process and the quality of the generated samples. Extensive experiments on multiple standard benchmarks demonstrate that ShortDF significantly reduces diffusion time (or steps) while enhancing the visual fidelity of generated samples compared to prior arts. This work, we suppose, paves the way for interactive diffusion-based applications and establishes a foundation for rapid data generation.

Built on trigular distribution, we learn an injective function mapping feature difference to label difference linearly. We prove it useful in 3 tasks: Facial Age Recognition, Illumination Chromaticity Estimation, and Aesthetics assessment.

The first awb work tacking awb problem in the world of the 3d point cloud, captured with rgb sensor and a TOF sensor.

To address the problem of decomposing an image into albedo and shading, we propose the Fast Fourier Intrinsic Network, FFI-Net in short, that operates in the spectral domain, splitting the input into several spectral bands.

We propose a neural network-based solution for three different tracks of 2nd International Illumination Estimation Challenge. SDE-AWB obtains 1st place in both indoor and two-illuminant tracks and 2nd place in general track.

Revisiting Temporal color constancy. We propose a 600-video temporal color constancy dataset, and a smaller better faster net called TCC-Net.

We cascade SqueezeNet-based bricks obtain SotA results on Gehlershi and NUS 8-camera datasets.

Gray (achromatic) pixels can be used for illumination estimation. This paper tells how to pick up them ACCURATELY.

Gray (achromatic) pixels can be used for illumination estimation. This paper tells how to pick up them using Clustering.

Gray (achromatic) pixels can be used for illumination estimation. This paper tells how to pick up them with the help of flash photography.

We briefly introduce two submissions to the Illumination Estimation Challenge, in the Int'l Workshop on Color Vision, affiliated to the 11th Int'l Symposium on Image and Signal Processing and Analysis. The fourier-transform-based submission is ranked 3rd, and the statistical Gray-pixel-based one ranked 6th.

Temporal color constancy: measuring the illumination of the captured image based on image deque.

Illumination estimation based on VGG and AlexNet feature map and multi-output regression. This is my first paper.