85 citations · 589 across the 37 of their papers we have counts for
37 papers
UniDiff: Advancing Vision-Language Models with Generative and Discriminative Learning
Xiao Dong, Runhui Huang, Xiaoyong Wei +4
Recent advances in vision-language pre-training have enabled machines to perform better in multimodal object discrimination (e.g., image-text semantic alignment) and image synthesi…
FastPillars: A Deployment-friendly Pillar-based 3D Detector
Sifan Zhou, Zhi Tian, Xiangxiang Chu +7
The deployment of 3D detectors strikes one of the major challenges in real-world self-driving scenarios. Existing BEV-based (i.e., Bird Eye View) detectors favor sparse convolution…
Multiple Object Tracking Challenge Technical Report for Team MT_IoT
Feng Yan, Zhiheng Li, Weixin Luo +4
This is a brief technical report of our proposed method for Multiple-Object Tracking (MOT) Challenge in Complex Environments. In this paper, we treat the MOT task as a two-stage ta…
AeDet: Azimuth-invariant Multi-view 3D Object Detection
Chengjian Feng, Zequn Jie, Yujie Zhong +2
Recent LSS-based multi-view 3D object detection has made tremendous progress, by processing the features in Brid-Eye-View (BEV) via the convolutional detector. However, the typical…
Expansion and Shrinkage of Localization for Weakly-Supervised Semantic Segmentation
Jinlong Li, Zequn Jie, Xu Wang +2
Generating precise class-aware pseudo ground-truths, a.k.a, class activation maps (CAMs), is essential for weakly-supervised semantic segmentation. The original CAM method usually…
Weakly Supervised Semantic Segmentation via Progressive Patch Learning
Jinlong Li, Zequn Jie, Xu Wang +3
Most of the existing semantic segmentation approaches with image-level class labels as supervision, highly rely on the initial class activation map (CAM) generated from the standar…
MSMDFusion: Fusing LiDAR and Camera at Multiple Scales with Multi-Depth Seeds for 3D Object Detection
Yang Jiao, Zequn Jie, Shaoxiang Chen +3
Fusing LiDAR and camera information is essential for achieving accurate and reliable 3D object detection in autonomous driving systems. This is challenging due to the difficulty of…
ARMANI: Part-level Garment-Text Alignment for Unified Cross-Modal Fashion Design
Xujie Zhang, Yu Sha, Michael C. Kampffmeyer +5
Cross-modal fashion image synthesis has emerged as one of the most promising directions in the generation domain due to the vast untapped potential of incorporating multiple modali…
MT-Net Submission to the Waymo 3D Detection Leaderboard
Shaoxiang Chen, Zequn Jie, Xiaolin Wei +1
In this technical report, we introduce our submission to the Waymo 3D Detection leaderboard. Our network is based on the Centerpoint architecture, but with significant improvements…
PromptDet: Towards Open-vocabulary Detection using Uncurated Images
Chengjian Feng, Yujie Zhong, Zequn Jie +5
The goal of this work is to establish a scalable pipeline for expanding an object detector towards novel/unseen categories, using zero manual annotations. To achieve that, we make…
MORE: Multi-Order RElation Mining for Dense Captioning in 3D Scenes
Yang Jiao, Shaoxiang Chen, Zequn Jie +3
3D dense captioning is a recently-proposed novel task, where point clouds contain more geometric information than the 2D counterpart. However, it is also more challenging due to th…
Suspected Object Matters: Rethinking Model's Prediction for One-stage Visual Grounding
Yang Jiao, Zequn Jie, Jingjing Chen +2
Recently, one-stage visual grounders attract high attention due to their comparable accuracy but significantly higher efficiency than two-stage grounders. However, inter-object rel…
Two-stage Visual Cues Enhancement Network for Referring Image Segmentation
Yang Jiao, Zequn Jie, Weixin Luo +4
Referring Image Segmentation (RIS) aims at segmenting the target object from an image referred by one given natural language expression. The diverse and flexible expressions as wel…
Delving into the Imbalance of Positive Proposals in Two-stage Object Detection
Zheng Ge, Zequn Jie, Xin Huang +2
Imbalance issue is a major yet unsolved bottleneck for the current object detection models. In this work, we observe two crucial yet never discussed imbalance issues. The first imb…
NMS by Representative Region: Towards Crowded Pedestrian Detection by Proposal Pairing
Xin Huang, Zheng Ge, Zequn Jie +1
Although significant progress has been made in pedestrian detection recently, pedestrian detection in crowded scenes is still challenging. The heavy occlusion between pedestrians i…
MTL-NAS: Task-Agnostic Neural Architecture Search towards General-Purpose Multi-Task Learning
Yuan Gao, Haoping Bai, Zequn Jie +3
We propose to incorporate neural architecture search (NAS) into general-purpose multi-task learning (GP-MTL). Existing NAS methods typically define different search spaces accordin…
PS-RCNN: Detecting Secondary Human Instances in a Crowd via Primary Object Suppression
Zheng Ge, Zequn Jie, Xin Huang +2
Detecting human bodies in highly crowded scenes is a challenging problem. Two main reasons result in such a problem: 1). weak visual cues of heavily occluded instances can hardly p…
Central Similarity Quantization for Efficient Image and Video Retrieval
Li Yuan, Tao Wang, Xiaopeng Zhang +4
Existing data-dependent hashing methods usually learn hash functions from pairwise or triplet data relationships, which only capture the data similarity locally, and often suffer f…
Real-Time Referring Expression Comprehension by Single-Stage Grounding Network
Xinpeng Chen, Lin Ma, Jingyuan Chen +3
In this paper, we propose a novel end-to-end model, namely Single-Stage Grounding network (SSG), to localize the referent given a referring expression within an image. Different fr…
A Sufficient Condition for Convergences of Adam and RMSProp
Fangyu Zou, Li Shen, Zequn Jie +2
Adam and RMSProp are two of the most influential adaptive stochastic algorithms for training deep neural networks, which have been pointed out to be divergent even in the convex se…
Modeling Varying Camera-IMU Time Offset in Optimization-Based Visual-Inertial Odometry
Yonggen Ling, Linchao Bao, Zequn Jie +6
Combining cameras and inertial measurement units (IMUs) has been proven effective in motion tracking, as these two sensing modalities offer complementary characteristics that are s…
A Unified Analysis of AdaGrad with Weighted Aggregation and Momentum Acceleration
Li Shen, Congliang Chen, Fangyu Zou +3
Integrating adaptive learning rate and momentum techniques into SGD leads to a large class of efficiently accelerated adaptive stochastic algorithms, such as AdaGrad, RMSProp, Adam…
Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation
Yunchao Wei, Huaxin Xiao, Honghui Shi +3
Despite the remarkable progress, weakly supervised segmentation approaches are still inferior to their fully supervised counterparts. We obverse the performance gap mainly comes fr…
Modular Generative Adversarial Networks
Bo Zhao, Bo Chang, Zequn Jie +1
Existing methods for multi-domain image-to-image translation (or generation) attempt to directly map an input image (or a random vector) to an image in one of the output domains. H…