Wei‐Lun Chao

The Ohio State University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Kilian Q. Weinberger

Laboratoire d'Informatique de Paris-Nord

Mark Campbell

Cornell University

Bharath Hariharan

Laboratoire d'Informatique de Paris-Nord

Fei Sha

Ocean University of China

Yurong You

Laboratoire d'Informatique de Paris-Nord

Hong-You Chen

Shanghai Municipal Center For Disease Control Prevention

Tai-Yu Pan

Shanghai Maritime University

Soravit Changpinyo

Google (United States)

Somashekar G. Krishna

The Ohio State University Wexner Medical Center

Tanya Berger‐Wolf

The Ohio State University

Cooperative Institutions

Weatherford College

406

Adrian College

334

Directorate of Medicinal and Aromatic Plants Research

330

The Ohio State University

114

Istituto Tecnico Industriale Alessandro Volta

111

Chinese Academy of Sciences

Google (United States)

The Ohio State University Wexner Medical Center

Cornell University

National Taiwan University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Deep Co-Training with Task Decomposition for Semi-Supervised Domain Adaptation

2021 IEEE/CVF International Conference on Computer Vision (ICCV) (2021)

Luyu Yang Yan Wang Mingfei Gao Abhinav Shrivastava Kilian Q. Weinberger

Semi-supervised domain adaptation (SSDA) aims to adapt models trained from a labeled source domain to a different but related target domain, from which unlabeled data and a small set of labeled data are provided. Current methods that treat source and target supervision without distinction overlook their inherent discrepancy, resulting in a source-dominated model that has not effectively use the target supervision. In this paper, we argue that the labeled target data needs to be distinguished for effective SSDA, and propose to explicitly decompose the SSDA task into two sub-tasks: a semi-supervised learning (SSL) task in the target domain and an unsupervised domain adaptation (UDA) task across domains. By doing so, the two sub-tasks can better leverage the corresponding supervision and thus yield very different classifiers. To integrate the strengths of the two classifiers, we apply the well established co-training framework, in which the two classifiers exchange their high confident predictions to iteratively "teach each other" so that both classifiers can excel in the target domain. We call our approach Deep Co-training with Task decomposition (DeCoTa). DeCoTa requires no adversarial training and is easy to implement. Moreover, DeCoTa is well founded on the theoretical condition of when co-training would succeed. As a result, DeCoTa achieves state-of-the-art results on several SSDA datasets, outperforming the prior art by a notable 4% margin on DomainNet. Code is available at https://github.com/LoyoYang/DeCoTa.

Domain Adaptation

Co-training

Leverage (statistics)

Labeled data

Margin (machine learning)

Training set

Task Analysis

10.1109/iccv48922.2021.00878

Cite

Citations (56)

Video Summarization with Long Short-term Memory

arXiv (Cornell University) (2016)

Ke Zhang Wei‐Lun Chao Fei Sha Kristen Grauman

We propose a novel supervised learning technique for summarizing videos by automatically selecting keyframes or key subshots. Casting the problem as a structured prediction problem on sequential data, our main idea is to use Long Short-Term Memory (LSTM), a special type of recurrent neural networks to model the variable-range dependencies entailed in the task of video summarization. Our learning models attain the state-of-the-art results on two benchmark video datasets. Detailed analysis justifies the design of the models. In particular, we show that it is crucial to take into consideration the sequential structures in videos and model them. Besides advances in modeling techniques, we introduce techniques to address the need of a large number of annotated data for training complex learning models. There, our main idea is to exploit the existence of auxiliary annotated video datasets, albeit heterogeneous in visual styles and contents. Specifically, we show domain adaptation techniques can improve summarization by reducing the discrepancies in statistical properties across those datasets.

Benchmark (surveying)

10.48550/arxiv.1605.08110

Cite

Citations (75)

Probabilistic Uncertainty Quantification of Prediction Models with Application to Visual Localization

arXiv (Cornell University) (2023)

Junan Chen Josephine Monica Wei‐Lun Chao Mark Campbell

The uncertainty quantification of prediction models (e.g., neural networks) is crucial for their adoption in many robotics applications. This is arguably as important as making accurate predictions, especially for safety-critical applications such as self-driving cars. This paper proposes our approach to uncertainty quantification in the context of visual localization for autonomous driving, where we predict locations from images. Our proposed framework estimates probabilistic uncertainty by creating a sensor error model that maps an internal output of the prediction model to the uncertainty. The sensor error model is created using multiple image databases of visual localization, each with ground-truth location. We demonstrate the accuracy of our uncertainty prediction framework using the Ithaca365 dataset, which includes variations in lighting, weather (sunny, snowy, night), and alignment errors between databases. We analyze both the predicted uncertainty and its incorporation into a Kalman-based localization filter. Our results show that prediction error variations increase with poor weather and lighting condition, leading to greater uncertainty and outliers, which can be predicted by our proposed uncertainty model. Additionally, our probabilistic error model enables the filter to remove ad hoc sensor gating, as the uncertainty automatically adjusts the model to the input data

Ground truth

Uncertainty Quantification

Errors-in-Variables Models

10.48550/arxiv.2305.20044

Cite

Citations (0)

Towards Open-World Segmentation of Parts

arXiv (Cornell University) (2023)

Tai-Yu Pan Qing Liu Wei‐Lun Chao Brian Price

Segmenting object parts such as cup handles and animal bodies is important in many real-world applications but requires more annotation effort. The largest dataset nowadays contains merely two hundred object categories, implying the difficulty to scale up part segmentation to an unconstrained setting. To address this, we propose to explore a seemingly simplified but empirically useful and scalable task, class-agnostic part segmentation. In this problem, we disregard the part class labels in training and instead treat all of them as a single part class. We argue and demonstrate that models trained without part classes can better localize parts and segment them on objects unseen in training. We then present two further improvements. First, we propose to make the model object-aware, leveraging the fact that parts are "compositions", whose extents are bounded by the corresponding objects and whose appearances are by nature not independent but bundled. Second, we introduce a novel approach to improve part segmentation on unseen objects, inspired by an interesting finding -- for unseen objects, the pixel-wise features extracted by the model often reveal high-quality part segments. To this end, we propose a novel self-supervised procedure that iterates between pixel clustering and supervised contrastive learning that pulls pixels closer or pushes them away. Via extensive experiments on PartImageNet and Pascal-Part, we show notable and consistent gains by our approach, essentially a critical step towards open-world part segmentation.

Pascal (unit)

Market Segmentation

Segmentation-based object categorization

10.48550/arxiv.2305.16804

Cite

Citations (0)

Pre-Training LiDAR-Based 3D Object Detectors Through Colorization

arXiv (Cornell University) (2023)

Tai-Yu Pan Chenyang Ma Tianle Chen Cheng Perng Phoo Katie Z Luo

Accurate 3D object detection and understanding for self-driving cars heavily relies on LiDAR point clouds, necessitating large amounts of labeled data to train. In this work, we introduce an innovative pre-training approach, Grounded Point Colorization (GPC), to bridge the gap between data and labels by teaching the model to colorize LiDAR point clouds, equipping it with valuable semantic cues. To tackle challenges arising from color variations and selection bias, we incorporate color as "context" by providing ground-truth colors as hints during colorization. Experimental results on the KITTI and Waymo datasets demonstrate GPC's remarkable effectiveness. Even with limited labeled data, GPC significantly improves fine-tuning performance; notably, on just 20% of the KITTI dataset, GPC outperforms training from scratch with the entire dataset. In sum, we introduce a fresh perspective on pre-training for 3D object detection, aligning the objective with the model's intended role and ultimately advancing the accuracy and efficiency of 3D object detection for autonomous vehicles.

Ground truth

Bridge (graph theory)

Training set

10.48550/arxiv.2310.14592

Cite

Citations (0)

Muscle injury determination by image segmentation

Jian–Jiun Ding Yu‐Hsiang Wang Lee-Lin Hu Wei‐Lun Chao Yio‐Wha Shau

Clinical examination of Congenital Muscular Torticollis (CMT) is often carried out by ultrasound equipments. However, a variety of subjective factors during diagnosis may result in wrong decision. Thus, we propose an image processing algorithm to derive the objective judgment on the healthiness of muscle in this paper. We first apply image segmentation technique, such as the fast scanning algorithm, for ultrasonic muscle image segmentation. Then, the proposed algorithms are applied to determine the healthiness of muscle fibers. We furthermore propose a score criterion to evaluate the degree of injury. The experimental results show that the injury score measured by the proposed methods can successfully determine whether the muscle is hurt and infer the extent of fibrosis.

Muscle fibre

10.1109/vcip.2011.6115925

Cite

Citations (3)

PreSTU: Pre-Training for Scene-Text Understanding

2021 IEEE/CVF International Conference on Computer Vision (ICCV) (2023)

Jihyung Kil Soravit Changpinyo Xi Chen Hexiang Hu Sebastian Goodman

The ability to recognize and reason about text embedded in visual inputs is often lacking in vision-and-language (V&L) models, perhaps because V&L pre-training methods have often failed to include such an ability in their training objective. In this paper, we propose PreSTU, a novel pre-training recipe dedicated to scene-text understanding (STU). PreSTU introduces OCR-aware pre-training objectives that encourage the model to recognize text from an image and connect it to the rest of the image content. We implement PreSTU using a simple transformer-based encoder-decoder architecture, combined with large-scale image-text datasets with scene text obtained from an off-the-shelf OCR system. We empirically demonstrate the effectiveness of this pre-training approach on eight visual question answering and four image captioning benchmarks.

Closed captioning

Training set

10.1109/iccv51070.2023.01401

Cite

Citations (8)

BioCLIP: A Vision Foundation Model for the Tree of Life

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2024)

Samuel Stevens Jiaman Wu Matthew J. Thompson Elizabeth G Campolongo Chan Hee Song

Foundation (evidence)

Tree (set theory)

10.1109/cvpr52733.2024.01836

Cite

Citations (8)

End-to-End Pseudo-LiDAR for Image-Based 3D Object Detection

arXiv (Cornell University) (2020)

Rui Qian Divyansh Garg Yan Wang Yurong You Serge Belongie

Reliable and accurate 3D object detection is a necessity for safe autonomous driving. Although LiDAR sensors can provide accurate 3D point cloud estimates of the environment, they are also prohibitively expensive for many settings. Recently, the introduction of pseudo-LiDAR (PL) has led to a drastic reduction in the accuracy gap between methods based on LiDAR sensors and those based on cheap stereo cameras. PL combines state-of-the-art deep neural networks for 3D depth estimation with those for 3D object detection by converting 2D depth map outputs to 3D point cloud inputs. However, so far these two networks have to be trained separately. In this paper, we introduce a new framework based on differentiable Change of Representation (CoR) modules that allow the entire PL pipeline to be trained end-to-end. The resulting framework is compatible with most state-of-the-art networks for both tasks and in combination with PointRCNN improves over PL consistently across all benchmarks -- yielding the highest entry on the KITTI image-based 3D object detection leaderboard at the time of submission. Our code will be made available at https://github.com/mileyan/pseudo-LiDAR_e2e.

Representation

Code (set theory)

10.48550/arxiv.2004.03080

Cite

Citations (7)

Effectiveness of Machine Learning in Predicting Orthodontic Tooth Extractions: A Multi-Institutional Study

Bioengineering (2024)

Lily E. Etemad John P. Heiner Abdullah Al Amin Tai‐Hsien Wu Wei‐Lun Chao

The study aimed to evaluate the effectiveness of machine learning in predicting whether orthodontic patients would require extraction or non-extraction treatment using data from two university datasets. A total of 1135 patients, with 297 from University 1 and 838 from University 2, were included during consecutive enrollment periods. The study identified 20 inputs including 9 clinical features and 11 cephalometric measurements based on previous research. Random forest (RF) models were used to make predictions for both institutions. The performance of each model was assessed using sensitivity (SEN), specificity (SPE), accuracy (ACC), and feature ranking. The model trained on the combined data from two universities demonstrated the highest performance, achieving 50% sensitivity, 97% specificity, and 85% accuracy. When cross-predicting, where the University 1 (U1) model was applied to the University 2 (U2) data and vice versa, there was a slight decrease in performance metrics (ranging from 0% to 20%). Maxillary and mandibular crowding were identified as the most significant features influencing extraction decisions in both institutions. This study is among the first to utilize datasets from two United States institutions, marking progress toward developing an artificial intelligence model to support orthodontists in clinical practice.

Crowding

Clinical Practice

10.3390/bioengineering11090888

Cite

Citations (1)