no code implementations • 7 Feb 2025 • Haoyu Han, Heng Yang
Global bundle adjustment is made easy by depth prediction and convex optimization.
no code implementations • 2 Feb 2025 • Han Qi, Haocheng Yin, Yilun Du, Heng Yang
Focusing on planar pushing with rich contact and collision, we show GPC dominates behavior cloning across state-based and vision-based, simulated and real-world experiments.
no code implementations • 10 Dec 2024 • Ziqi Lu, Heng Yang, Danfei Xu, Boyi Li, Boris Ivanovic, Marco Pavone, Yue Wang
Emerging 3D geometric foundation models, such as DUSt3R, offer a promising approach for in-the-wild 3D vision tasks.
no code implementations • 25 Nov 2024 • Zequn Chen, Jiezhi Yang, Heng Yang
We present PreF3R, Pose-Free Feed-forward 3D Reconstruction from an image sequence of variable length.
1 code implementation • 4 Nov 2024 • Elad Sharony, Heng Yang, Tong Che, Marco Pavone, Shie Mannor, Peter Karkus
Sequentially solving similar optimization problems under strict runtime constraints is essential for many applications, such as robot control, autonomous driving, and portfolio management.
no code implementations • 16 Oct 2024 • Minkyoung Cho, Yulong Cao, Jiachen Sun, Qingzhao Zhang, Marco Pavone, Jeong Joon Park, Heng Yang, Z. Morley Mao
An important paradigm in 3D object detection is the use of multiple modalities to enhance accuracy in both normal and challenging conditions, particularly for long-tail scenarios.
no code implementations • 7 Oct 2024 • Han Qi, Haocheng Yin, Heng Yang
Surprisingly, such an NC-pretrained vision encoder, when finetuned end-to-end with the action decoder, boosts the test-time performance by 10% to 35%.
1 code implementation • 3 Oct 2024 • ZhiYu Zhang, Zhou Lu, Heng Yang
By converting the target confidence levels into quantile levels, the problem can be reduced to predicting the quantiles (in hindsight) of a sequentially revealed data sequence.
2 code implementations • 2 Oct 2024 • Heng Yang, Jack Cole, Ke Li
Recent breakthroughs in GFMs, such as Evo, have attracted significant investment and attention to genomic modeling, as they address long-standing challenges and transform in-silico genomic studies into automated, reliable, and efficient paradigms.
1 code implementation • 15 Jul 2024 • Heng Yang, Renzhi Chen, Ke Li
The alignment between RNA sequences and structures in foundation models (FMs) has yet to be thoroughly investigated.
no code implementations • 3 Jun 2024 • Kevin Kasa, ZhiYu Zhang, Heng Yang, Graham W. Taylor
Conformal prediction (CP) enables machine learning models to output prediction sets with guaranteed coverage rate, assuming exchangeable data.
no code implementations • 26 May 2024 • Aneesh Muppidi, ZhiYu Zhang, Heng Yang
A key challenge in lifelong reinforcement learning (RL) is the loss of plasticity, where previous learning progress hinders an agent's adaptation to new tasks.
no code implementations • 12 Mar 2024 • Chensheng Peng, Chenfeng Xu, Yue Wang, Mingyu Ding, Heng Yang, Masayoshi Tomizuka, Kurt Keutzer, Marco Pavone, Wei Zhan
In this paper, we reimagine volumetric representations through the lens of quadrics.
1 code implementation • 5 Feb 2024 • ZhiYu Zhang, David Bombara, Heng Yang
We study online learning in adversarial nonstationary environments.
no code implementations • NeurIPS 2023 • Apoorva Sharma, Sushant Veer, Asher Hancock, Heng Yang, Marco Pavone, Anirudha Majumdar
To remedy this, recent work has proposed learning model and score function parameters using data to directly optimize the efficiency of the ICP prediction sets.
1 code implementation • 26 Oct 2023 • Heng Yang, Ke Li
Instruction-based language modeling has received significant attention in pretrained language models.
no code implementations • 27 Sep 2023 • ZhiYu Zhang, Heng Yang, Ashok Cutkosky, Ioannis Ch. Paschalidis
Motivated by the pursuit of instance optimality, we propose a new algorithm that simultaneously achieves ($i$) the AdaGrad-style second order gradient adaptivity; and ($ii$) the comparator norm adaptivity also known as "parameter freeness" in the literature.
no code implementations • 30 Jun 2023 • Ping Zhang, Heng Yang, Zhiyong Feng, Yanpeng Cui, Jincheng Dai, Xiaoqi Qin, Jinglin Li, Qixun Zhang
Driven by the vision of "intelligent connection of everything" toward 6G, the collective intelligence of networked machines can be fully exploited to improve system efficiency by shifting the paradigm of wireless communication design from naive maximalist approaches to intelligent value-based approaches.
no code implementations • 6 May 2023 • Heng Yang, Ke Li
Recent studies have revealed the vulnerability of pre-trained language models to adversarial attacks.
1 code implementation • CVPR 2023 • Heng Yang, Marco Pavone
Geometric uncertainty propagation, on the other, propagates the geometric constraints on the keypoints to the 6D object pose, leading to a Pose UnceRtainty SEt (PURSE) that guarantees coverage of the groundtruth pose with the same probability.
no code implementations • 24 Jan 2023 • Xiao He, Mingrui Zhu, Nannan Wang, Xinbo Gao, Heng Yang
To address this issue, we propose a novel font generation approach by learning the Difference between different styles and the Similarity of the same style (DS-Font).
1 code implementation • NIPS 2022 • De Cheng, Yixiong Ning, Nannan Wang, Xinbo Gao, Heng Yang, Yuxuan Du, Bo Han, Tongliang Liu
We show that the cycle-consistency regularization helps to minimize the volume of the transition matrix T indirectly without exploiting the estimated noisy class posterior, which could further encourage the estimated transition matrix T to converge to its optimal solution.
1 code implementation • 6 Oct 2022 • Heng Yang, Ke Li
Our experimental results on three classification tasks and nine public datasets show that BootAug addresses the performance drop problem and outperforms state-of-the-art text augmentation methods.
2 code implementations • 2 Aug 2022 • Heng Yang, Chen Zhang, Ke Li
The advancement of aspect-based sentiment analysis (ABSA) has urged the lack of a user-friendly framework that can largely lower the difficulty of reproducing state-of-the-art ABSA performance, especially for beginners.
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+5
1 code implementation • 24 Jun 2022 • Jingnan Shi, Heng Yang, Luca Carlone
We consider an active shape model, where -- for an object category -- we are given a library of potential CAD models describing objects in that category, and we adopt a standard formulation where pose and shape are estimated from 2D or 3D keypoints via non-convex optimization.
1 code implementation • 12 Jun 2022 • Ke Li, Heng Yang, Willem Visser
In this paper, we propose DaNuoYi, an automatic injection testing tool that simultaneously generates test inputs for multiple types of injection attacks on a WAF.
1 code implementation • journal 2022 • Mayi Xu, Biqing Zeng, Heng Yang, Junlong Chi, Jiatao Chen, Hongye Liu a
The DLCF can dynamically capture the range of local context based on the different max distance from the target aspect term to its context words and the DCA allows the model to pay more attention to the cluster which is more critical for sentiment classification.
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+2
1 code implementation • 16 Oct 2021 • Heng Yang, Ke Li
Aspect sentiment coherency is an intriguing yet underexplored topic in the field of aspect-based sentiment classification.
Adversarial Defense
Aspect-Based Sentiment Analysis (ABSA)
+2
2 code implementations • 7 Sep 2021 • Heng Yang, Luca Carlone
Our third contribution is to solve the SDP relaxations at an unprecedented scale and accuracy by presenting STRIDE, a solver that blends global descent on the convex SDP with fast local search on the nonconvex POP.
1 code implementation • 28 May 2021 • Heng Yang, Ling Liang, Luca Carlone, Kim-Chuan Toh
In particular, we first design a globally convergent inexact projected gradient method (iPGM) for solving the SDP that serves as the backbone of our framework.
1 code implementation • 16 Apr 2021 • Jingnan Shi, Heng Yang, Luca Carlone
Our first contribution is to provide the first certifiably optimal solver for pose and shape estimation.
1 code implementation • ICCV 2021 • Heng Yang, Chris Doran, Jean-Jacques Slotine
We study the problem of aligning two sets of 3D geometric primitives given known correspondences.
2 code implementations • CVPR 2021 • Heng Yang, Wei Dong, Luca Carlone, Vladlen Koltun
We present self-supervised geometric perception (SGP), the first general framework to learn a feature descriptor for correspondence matching without any ground-truth geometric model labels (e. g., camera poses, rigid transformations).
2 code implementations • 7 Nov 2020 • Jingnan Shi, Heng Yang, Luca Carlone
We also show that in practice the maximum k-core of the compatibility graph provides an approximation of the maximum clique, while being faster to compute in large problems.
2 code implementations • 2 Oct 2020 • Heng Yang, Biqing Zeng
Target-oriented sentiment classification is a fine-grained task of natural language processing to analyze the sentiment polarity of the targets.
no code implementations • 29 Jul 2020 • Pasquale Antonante, Vasileios Tzoumas, Heng Yang, Luca Carlone
We extend ADAPT and GNC to the case where the user does not have prior knowledge of the inlier-noise statistics (or the statistics may vary over time) and is unable to guess a reasonable threshold to separate inliers from outliers (as the one commonly used in RANSAC).
1 code implementation • NeurIPS 2020 • Heng Yang, Luca Carlone
We propose the first general and practical framework to design certifiable algorithms for robust geometric perception in the presence of a large amount of outliers.
no code implementations • 7 May 2020 • Heng Yang
We provide a dynamical perspective on the classical problem of 3D point cloud registration with correspondences.
7 code implementations • 21 Jan 2020 • Heng Yang, Jingnan Shi, Luca Carlone
We propose the first fast and certifiable algorithm for the registration of two sets of 3D points in the presence of large amounts of outlier correspondences.
6 code implementations • 17 Dec 2019 • Heng Yang, Biqing Zeng, JianHao Yang, Youwei Song, Ruyang Xu
Aspect-based sentiment analysis (ABSA) task is a multi-grained task of natural language processing and consists of two subtasks: aspect term extraction (ATE) and aspect polarity classification (APC).
Ranked #5 on
Aspect-Based Sentiment Analysis (ABSA)
on SemEval-2014 Task-4
(using extra training data)
Aspect-Based Sentiment Analysis
Aspect-Based Sentiment Analysis (ABSA)
+3
no code implementations • CVPR 2020 • Heng Yang, Luca Carlone
We study the problem of 3D shape reconstruction from 2D landmarks extracted in a single image.
4 code implementations • 18 Sep 2019 • Heng Yang, Pasquale Antonante, Vasileios Tzoumas, Luca Carlone
In this paper, we enable the simultaneous use of non-minimal solvers and robust estimation by providing a general-purpose approach for robust global estimation, which can be applied to any problem where a non-minimal solver is available for the outlier-free case.
4 code implementations • ICCV 2019 • Heng Yang, Luca Carlone
Our first contribution is to formulate the Wahba problem using a Truncated Least Squares (TLS) cost that is insensitive to a large fraction of spurious correspondences.
2 code implementations • 20 Mar 2019 • Heng Yang, Luca Carlone
We propose a robust approach for the registration of two sets of 3D points in the presence of a large amount of outliers.
no code implementations • 14 Jun 2018 • Huiyuan Zhuo, Xuelin Qian, Yanwei Fu, Heng Yang, xiangyang xue
In this paper, we proposed a novel filter pruning for convolutional neural networks compression, namely spectral clustering filter pruning with soft self-adaption manners (SCSP).
no code implementations • 17 Apr 2018 • Jiaolong Xu, Peng Wang, Heng Yang, Antonio M. López
Autonomous driving has harsh requirements of small model size and energy efficiency, in order to enable the embedded system to achieve real-time on-board object detection.
no code implementations • 16 Nov 2015 • Heng Yang, Xuhui Jia, Chen Change Loy, Peter Robinson
In this paper, we carry out a rigorous evaluation of these methods by making the following contributions: 1) we proposes a new evaluation metric for face alignment on a set of images, i. e., area under error distribution curve within a threshold, AUC$_\alpha$, given the fact that the traditional evaluation measure (mean error) is very sensitive to big alignment error.
no code implementations • 16 Sep 2015 • Heng Yang, Renqiao Zhang, Peter Robinson
Furthermore, we study the impact of training data imbalance on model performance and propose a training sample augmentation scheme that produces more initialisations for training samples from the minority.
1 code implementation • 11 Jul 2015 • Heng Yang, Wenxuan Mou, Yichi Zhang, Ioannis Patras, Hatice Gunes, Peter Robinson
In this paper we propose a supervised initialization scheme for cascaded face alignment based on explicit head pose estimation.
no code implementations • CVPR 2015 • Heng Yang, Ioannis Patras
Our experiments lead to several interesting findings: 1) Surprisingly, most of state of the art methods struggle to preserve the mirror symmetry, despite the fact that they do have very similar overall performance on the original and mirror images; 2) the low mirrorability is not caused by training or testing sample bias - all algorithms are trained on both the original images and their mirrored versions; 3) the mirror error is strongly correlated to the localization/alignment error (with correlation coefficients around 0. 7).
no code implementations • CVPR 2014 • Changqing Zou, Heng Yang, Jianzhuang Liu
Reconstructing 3D objects from single line drawings is often desirable in computer vision and graphics applications.