210 citations · 413 across the 18 of their papers we have counts for
18 papers
MOFI: Learning Image Representations from Noisy Entity Annotated Images
Wentao Wu, Aleksei Timofeev, Chen Chen +8
We present MOFI, Manifold OF Images, a new vision foundation model designed to learn image representations from noisy entity annotated images. MOFI differs from previous work in tw…
Less is More: Removing Text-regions Improves CLIP Training Efficiency and Robustness
Liangliang Cao, Bowen Zhang, Chen Chen +5
The CLIP (Contrastive Language-Image Pre-training) model and its variants are becoming the de facto backbone in many applications. However, training a CLIP model from hundreds of m…
Optimizing Anchor-based Detectors for Autonomous Driving Scenes
Xianzhi Du, Wei-Chih Hung, Tsung-Yi Lin
This paper summarizes model improvements and inference-time optimizations for the popular anchor-based detectors in the scenes of autonomous driving. Based on the high-performing R…
Auto-scaling Vision Transformers without Training
Wuyang Chen, Wei Huang, Xianzhi Du +3
This work targets automated designing and scaling of Vision Transformers (ViTs). The motivation comes from two pain spots: 1) the lack of efficient and principled methods for desig…
Provable Stochastic Optimization for Global Contrastive Learning: Small Batch Does Not Harm Performance
Zhuoning Yuan, Yuexin Wu, Zi-Hao Qiu +4
In this paper, we study contrastive learning from an optimization perspective, aiming to analyze and address a fundamental issue of existing contrastive learning methods that eithe…
Towards a Unified Foundation Model: Jointly Pre-Training Transformers on Unpaired Images and Text
Qing Li, Boqing Gong, Yin Cui +4
In this paper, we explore the possibility of building a unified foundation model that can be adapted to both vision-only and text-only tasks. Starting from BERT and ViT, we design…
A Simple Single-Scale Vision Transformer for Object Localization and Instance Segmentation
Wuyang Chen, Xianzhi Du, Fan Yang +8
This work presents a simple vision transformer design as a strong baseline for object localization and instance segmentation tasks. Transformers recently demonstrate competitive pe…
Revisiting 3D ResNets for Video Recognition
Xianzhi Du, Yeqing Li, Yin Cui +3
A recent work from Bello shows that training and scaling strategies may be more significant than model architectures for visual recognition. This short note studies effective train…
Simple Training Strategies and Model Scaling for Object Detection
Xianzhi Du, Barret Zoph, Wei-Chih Hung +1
The speed-accuracy Pareto curve of object detection systems have advanced through a combination of better model architectures, training and inference methods. In this paper, we met…
Dilated SpineNet for Semantic Segmentation
Abdullah Rashwan, Xianzhi Du, Xiaoqi Yin +1
Scale-permuted networks have shown promising results on object bounding box detection and instance segmentation. Scale permutation and cross-scale fusion of features enable the net…
Revisiting ResNets: Improved Training and Scaling Strategies
Irwan Bello, William Fedus, Xianzhi Du +5
Novel computer vision architectures monopolize the spotlight, but the impact of the model architecture is often conflated with simultaneous changes to training methodology and scal…
Efficient Scale-Permuted Backbone with Learned Resource Distribution
Xianzhi Du, Tsung-Yi Lin, Pengchong Jin +4
Recently, SpineNet has demonstrated promising results on object detection and image classification over ResNet model. However, it is unclear if the improvement adds up when combini…
SpineNet: Learning Scale-Permuted Backbone for Recognition and Localization
Xianzhi Du, Tsung-Yi Lin, Pengchong Jin +5
Convolutional neural networks typically encode an input image into a series of intermediate features with decreasing resolutions. While this structure is suited to classification t…
TW-SMNet: Deep Multitask Learning of Tele-Wide Stereo Matching
Mostafa El-Khamy, Haoyu Ren, Xianzhi Du +1
In this paper, we introduce the problem of estimating the real world depth of elements in a scene captured by two cameras with different field of views, where the first field of vi…
AMNet: Deep Atrous Multiscale Stereo Disparity Estimation Networks
Xianzhi Du, Mostafa El-Khamy, Jungwon Lee
In this paper, a new deep learning architecture for stereo disparity estimation is proposed. The proposed atrous multiscale network (AMNet) adopts an efficient feature extractor wi…
Fused Deep Neural Networks for Efficient Pedestrian Detection
Xianzhi Du, Mostafa El-Khamy, Vlad I. Morariu +2
In this paper, we present an efficient pedestrian detection system, designed by fusion of multiple deep neural network (DNN) systems. Pedestrian candidates are first generated by a…
Boundary-sensitive Network for Portrait Segmentation
Xianzhi Du, Xiaolong Wang, Dawei Li +5
Compared to the general semantic segmentation problem, portrait segmentation has higher precision requirement on boundary area. However, this problem has not been well studied in p…
Fused DNN: A deep neural network fusion approach to fast and robust pedestrian detection
Xianzhi Du, Mostafa El-Khamy, Jungwon Lee +1
We propose a deep neural network fusion architecture for fast and robust pedestrian detection. The proposed network fusion architecture allows for parallel processing of multiple n…