Shen Zheng 「郑深」

I am a second-year PhD student at the Robotics Institute, Carnegie Mellon University (CMU), advised by Dr. Srinivasa Narasimhan. I have interned at Waymo and Momenta, and worked at Lucid Motors. I completed my M.S. in Computer Vision (MSCV) at CMU, and earned my B.S. in Mathematics from Wenzhou-Kean University (WKU), where I worked with Dr. Gaurav Gupta.

Email1: shenzhen@andrew.cmu.edu

Email2: lebronshenzheng@gmail.com

Resume  /  Google Scholar  /  Github  /  Leetcode

My current research focuses on semantic-aware image warping and contrastive learning (e.g., WarpI2I, Instance-Warp, TPSeNCE) to enhance visual details in generative models and improve the understanding of small and long-tail objects (e.g., ROADWork) in discriminative models.

My earlier works focused on image restoration and enhancement (e.g., SGZ, LLIE_Survey).

Motivation: Low-capacity image-to-image translation models often struggle to preserve fine details because they compress high-resolution inputs into a spatially downsampled latent space.

Solution: We warp the input image to enlarge salient regions (e.g., objects, faces, eyes) so they are better preserved during image-to-image translation in the compressed latent representation.

Motivation: Navigating work zones is challenging due to the lack of large-scale datasets capturing their complexity and variability.

Solution: We introduce ROADWork, a large-scale open-source dataset and benchmark with fine-grained annotations and scene descriptions to support robust perception and navigation in complex construction zones.

Motivation: In domain adaptation, backgrounds occupy more pixels and exhibit larger cross-domain variations, making them harder to adapt, while foreground objects occupy fewer pixels and are less diverse, thus easier to adapt.

Solution: We warp the image at instance-level to oversample easier-to-adapt foreground objects and undersample harder-to-adapt background regions to improve feature representation in domain adaptation.

Motivation: Previous image-to-image translation methods produce artifacts and distortions, and lack control over the amount of rain generated.

Solution: Introduce a Triangular Probability Similarity (TPS) loss to minimize the artifacts and distortions. Propose a Semantic Noise Contrastive Estimation (SeNCE) strategy to stabilize the amounts of generated rain. Show that realistic rain generation benefits deraining and object detection in rain.

Motivation: Existing LLIE datasets focus on either overexposure or underexposure, not both, and usually feature minimally degraded images captured from static positions.

Solution: Present a comprehensive survey of low-light image enhancement (LLIE). Propose the SICE_Grad and SICE_Mix image datasets, which include images with both overexposure and underexposure. Introduce Night Wenzhou, a large-scale, high-resolution video dataset captured in fast motion with diverse illuminations and degradation.

Motivation: Current point cloud analysis methods struggle with irregular (i.e., unevenly distributed) point clouds.

Solution: PointNorm, a point cloud analysis network with a DualNorm module (Point Normalization & Reverse Point Normalization) that leverages local mean and global standard deviation.

Motivation: Current low-light image enhancement methods cannot handle uneven illuminations, is computationally inefficient, and fail to preserve the semantic information.

Solution: Introduce SGZ, a zero-shot low-light image enhancement framework with pixel-wise light deficiency estimation, parameter-free recurrent image enhancement, and unsupervised semantic segmentation.

Motivation: Generative models experience posterior collapse and vanishing gradient due to no effective metric for real-fake image evaluation.

Solution: Propose Adversarial Similarity Distance Introspective Variational Autoencoder (AS-IntroVAE), which can address the posterior collapse and the vanishing gradient problem in image generation in one go.

Mentor: Guohao Zhang

Improved Online HD Map Construction using long-term and short-term memory fusion.

Director: Dr. Feng Guo

Working as a perception software engineer in the ADAS perception team responsible for ADAS parking, traffic light detection, and blockage detection.

Director: Dr. Wangjiang Zhu

Responsible for long-tailed data augmentation, training data auto-labeling and cleaning, and model evaluation for traffic light detection algorithms.