Weizhe Liu | EPFL Graph Search

Are you an EPFL student looking for a semester project?

Work with us on data science and visualisation projects, and deploy your project as an app on top of GraphSearch.

I am a Ph.D. candidate at CVLAB, EPFL, under the supervision of Prof. Pascal Fua. Prior to that, I received the Master of Science degree from EPFL in 2017 and the Bachelor of Engineering degree from University of Electronic Science and Technology of China (UESTC) in 2014. My research interests lie in the field of Computer Vision, Machine Learning and Robotics. My main focus is on developing algorithms to crowd analysis problem, including counting, localization and motion estimation. Recently, I also work on semantic segmentation, domain adaptation and learning with less supervision.

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related units (2)

Related units

Courses taught by this person

Related research domains (8)

Related research domains

Related publications (10)

Related publications

People doing similar research (118)

People doing similar research

Deep learning

Deep learning is part of a broader family of machine learning methods, which is based on artificial neural networks with representation learning. The adjective "deep" in deep learning refers to the

Image segmentation

In and computer vision, image segmentation is the process of partitioning a into multiple image segments, also known as image regions or image objects (sets of pixels). The goal of segmentation is

Deep belief network

In machine learning, a deep belief network (DBN) is a generative graphical model, or alternatively a class of deep neural network, composed of multiple layers of latent variables ("hidden units"),

Multi-view Tracking Using Weakly Supervised Human Motion Prediction

Martin Pierre Engilberge, Pascal Fua, Weizhe Liu

Multi-view approaches to people-tracking have the potential to better handle occlusions than single-view ones in crowded scenes. They often rely on the tracking-by-detection paradigm, which involves detecting people first and then connecting the detections. In this paper, we argue that an even more effective approach is to predict people motion over time and infer people’s presence in individual frames from these. This enables to enforce consistency both over time and across views of a single temporal frame. We validate our approach on the PETS2009 and WILDTRACK datasets and demonstrate that it outperforms state-of-the-art methods.

2023

Learning to Align Sequential Actions in the Wild

Pascal Fua, Weizhe Liu, Bugra Tekin

State-of-the-art methods for self-supervised sequential action alignment rely on deep networks that find correspon- dences across videos in time. They either learn frame-to- frame mapping across sequences, which does not leverage temporal information, or assume monotonic alignment be- tween each video pair, which ignores variations in the or- der of actions. As such, these methods are not able to deal with common real-world scenarios that involve background frames or videos that contain non-monotonic sequence of actions. In this paper, we propose an approach to align sequential actions in the wild that involve diverse temporal variations. To this end, we propose an approach to enforce tempo- ral priors on the optimal transport matrix, which leverages temporal consistency, while allowing for variations in the order of actions. Our model accounts for both monotonic and non-monotonic sequences and handles background frames that should not be aligned. We demonstrate that our approach consistently outperforms the state-of-the-art in self-supervised sequential action representation learning on four different benchmark dataset.

IEEE2022

Leveraging Self-Supervision for Cross-Domain Crowd Counting

Nikita Durasov, Pascal Fua, Weizhe Liu

State-of-the-art methods for counting people in crowded scenes rely on deep networks to estimate crowd density. While effective, these data-driven approaches rely on large amount of data annotation to achieve good performance, which stops these models from being deployed in emergencies during which data annotation is either too costly or cannot be obtained fast enough. One popular solution is to use synthetic data for training. Unfortunately, due to domain shift, the resulting models generalize poorly on real imagery. We remedy this shortcoming by training with both synthetic images, along with their associated labels, and unlabeled real images. To this end, we force our network to learn perspective-aware features by training it to recognize upside-down real images from regular ones and incorporate into it the ability to predict its own uncertainty so that it can generate useful pseudo labels for fine-tuning purposes. This yields an algorithm that consistently outperforms state-of-the-art cross-domain crowd counting ones without any extra computation at inference time. Code is publicly available at https://github.com/weizheliu/Cross-Domain-Crowd-Counting.

IEEE COMPUTER SOC2022