no code implementations • 11 Dec 2023 • Ava Pun, Gary Sun, Jingkang Wang, Yun Chen, Ze Yang, Sivabalan Manivasagam, Wei-Chiu Ma, Raquel Urtasun
Different outdoor illumination conditions drastically alter the appearance of urban scenes, and they can harm the performance of image-based robot perception systems if not seen during training.
no code implementations • ICCV 2023 • Jeffrey Yunfan Liu, Yun Chen, Ze Yang, Jingkang Wang, Sivabalan Manivasagam, Raquel Urtasun
We propose a new method for realistic real-time novel-view synthesis (NVS) of large scenes.
no code implementations • 9 Nov 2023 • Ze Yang, Sivabalan Manivasagam, Yun Chen, Jingkang Wang, Rui Hu, Raquel Urtasun
In this work, we present NeuSim, a novel approach that estimates accurate geometry and realistic appearance from sparse in-the-wild data captured at distance and at limited viewpoints.
no code implementations • 2 Nov 2023 • Jingkang Wang, Sivabalan Manivasagam, Yun Chen, Ze Yang, Ioan Andrei Bârsan, Anqi Joyce Yang, Wei-Chiu Ma, Raquel Urtasun
To tackle these issues, we present CADSim, which combines part-aware object-class priors via a small set of CAD models with differentiable rendering to automatically reconstruct vehicle geometry, including articulated wheels, with high-quality appearance.
no code implementations • 2 Nov 2023 • Lunjun Zhang, Yuwen Xiong, Ze Yang, Sergio Casas, Rui Hu, Raquel Urtasun
Learning world models can teach an agent how the world works in an unsupervised manner.
no code implementations • CVPR 2023 • Ze Yang, Yun Chen, Jingkang Wang, Sivabalan Manivasagam, Wei-Chiu Ma, Anqi Joyce Yang, Raquel Urtasun
Previously recorded driving logs provide a rich resource to build these new scenarios from, but for closed loop evaluation, we need to modify the sensor data based on the new scene configuration and the SDV's decisions, as actors might be added or removed and the trajectories of existing actors and the SDV will differ from the original log.
no code implementations • ICCV 2023 • Sivabalan Manivasagam, Ioan Andrei Bârsan, Jingkang Wang, Ze Yang, Raquel Urtasun
We leverage this setting to analyze what aspects of LiDAR simulation, such as pulse phenomena, scanning effects, and asset quality, affect the domain gap with respect to the autonomy system, including perception, prediction, and motion planning, and analyze how modifications to the simulated LiDAR influence each part.
1 code implementation • ICCV 2023 • Ze Yang, Ruibo Li, Evan Ling, Chi Zhang, Yiming Wang, Dezhao Huang, Keng Teck Ma, Minhoe Hur, Guosheng Lin
To address this issue, we propose a new label-guided knowledge distillation (LGKD) loss, where the old model output is expanded and transplanted (with the guidance of the ground truth label) to form a semantically appropriate class correspondence with the new model output.
Ranked #1 on Continual Semantic Segmentation on ScanNet
no code implementations • Findings (NAACL) 2022 • Ze Yang, Liran Wang, Zhoujin Tian, Wei Wu, Zhoujun Li
Another is that applying the existing pre-trained models to this task is tricky because of the structural dependence within the conversation and its informal expression, etc.
1 code implementation • CVPR 2022 • Haiyang Wang, Shaoshuai Shi, Ze Yang, Rongyao Fang, Qi Qian, Hongsheng Li, Bernt Schiele, LiWei Wang
In order to learn better representations of object shape to enhance cluster features for predicting 3D boxes, we propose a ray-based feature grouping module, which aggregates the point-wise features on object surfaces using a group of determined rays uniformly emitted from cluster centers.
Ranked #13 on 3D Object Detection on ScanNetV2
1 code implementation • 23 Mar 2022 • Ze Yang, Chi Zhang, Ruibo Li, Yi Xu, Guosheng Lin
Upon this baseline, we devise an initializer named knowledge inheritance (KI) to reliably initialize the novel weights for the box classifier, which effectively facilitates the knowledge transfer process and boosts the adaptation speed.
no code implementations • 28 Jul 2021 • Ze Yang, Haofei Wang, Feng Lu
We evaluate the performance of three deep learning-based methods (Deepphys, rPPGNet, and Physnet) to that of four traditional methods (CHROM, GREEN, ICA, and POS) using two public datasets: the UBFC-rPPG dataset and the BH-rPPG dataset.
no code implementations • CVPR 2021 • Ze Yang, Shenlong Wang, Sivabalan Manivasagam, Zeng Huang, Wei-Chiu Ma, Xinchen Yan, Ersin Yumer, Raquel Urtasun
Constructing and animating humans is an important component for building virtual worlds in a wide variety of applications such as virtual reality or robotics testing in simulation.
no code implementations • 16 Nov 2020 • Ze Yang, Siva Manivasagam, Ming Liang, Bin Yang, Wei-Chiu Ma, Raquel Urtasun
We then incorporate the reconstructed pedestrian assets bank in a realistic LiDAR simulation system by performing motion retargeting, and show that the simulated LiDAR data can be used to significantly reduce the amount of annotated real-world data required for visual perception tasks.
1 code implementation • Findings of the Association for Computational Linguistics 2020 • Ze Yang, Wei Wu, Can Xu, Xinnian Liang, Jiaqi Bai, Liran Wang, Wei Wang, Zhoujun Li
Generating responses following a desired style has great potentials to extend applications of open-domain dialogue systems, yet is refrained by lacking of parallel data for training.
no code implementations • 4 Apr 2020 • Ze Yang, Wei Wu, Huang Hu, Can Xu, Wei Wang, Zhoujun Li
Thus, we propose learning a response generation model with both image-grounded dialogues and textual dialogues by assuming that the visual scene information at the time of a conversation can be represented by an image, and trying to recover the latent images of the textual dialogues through text-to-image generation techniques.
1 code implementation • 16 Mar 2020 • Ze Yang, Yali Wang, Xianyu Chen, Jianzhuang Liu, Yu Qiao
Few-shot object detection is a challenging but realistic scenario, where only a few annotated training images are available for training detectors.
2 code implementations • ECCV 2020 • Ze Yang, Yinghao Xu, Han Xue, Zheng Zhang, Raquel Urtasun, Li-Wei Wang, Stephen Lin, Han Hu
We present a new object representation, called Dense RepPoints, that utilizes a large set of points to describe an object at multiple levels, including both box level and pixel level.
no code implementations • 18 Oct 2019 • Ze Yang, Linjun Shou, Ming Gong, Wutao Lin, Daxin Jiang
The experiment results show that our method can significantly outperform the baseline methods and even achieve comparable results with the original teacher models, along with substantial speedup of model inference.
no code implementations • ICCV 2019 • Ze Yang, Liwei Wang
Recognizing 3D object has attracted plenty of attention recently, and view-based methods have achieved best results until now.
no code implementations • EMNLP2019 2019 • Ze Yang, Can Xu, Wei Wu, Zhoujun Li
Automatic news comment generation is a new testbed for techniques of natural language generation.
no code implementations • 27 Sep 2019 • Jinchen Xuan, Yunchang Yang, Ze Yang, Di He, Li-Wei Wang
Motivated by this observation, we discover two specific problems of GANs leading to anomalous generalization behaviour, which we refer to as the sample insufficiency and the pixel-wise combination.
no code implementations • IJCNLP 2019 • Ze Yang, Can Xu, Wei Wu, Zhoujun Li
Automatic news comment generation is a new testbed for techniques of natural language generation.
1 code implementation • IJCNLP 2019 • Ze Yang, Wei Wu, Jian Yang, Can Xu, Zhoujun Li
Since the paired data now is no longer enough to train a neural generation model, we consider leveraging the large scale of unpaired data that are much easier to obtain, and propose response generation with both paired and unpaired data.
6 code implementations • ICCV 2019 • Ze Yang, Shaohui Liu, Han Hu, Li-Wei Wang, Stephen Lin
They furthermore do not require the use of anchors to sample a space of bounding boxes.
Ranked #95 on Object Detection on COCO minival
2 code implementations • IJCNLP 2019 • Ming Gong, Linjun Shou, Wutao Lin, Zhijie Sang, Quanjia Yan, Ze Yang, Feixiang Cheng, Daxin Jiang
Deep Neural Networks (DNN) have been widely employed in industry to address various Natural Language Processing (NLP) tasks.
no code implementations • 21 Apr 2019 • Ze Yang, Linjun Shou, Ming Gong, Wutao Lin, Daxin Jiang
Deep pre-training and fine-tuning models (like BERT, OpenAI GPT) have demonstrated excellent results in question answering areas.
12 code implementations • ECCV 2018 • Ze Yang, Tiange Luo, Dong Wang, Zhiqiang Hu, Jun Gao, Li-Wei Wang
In consideration of intrinsic consistency between informativeness of the regions and their probability being ground-truth class, we design a novel training paradigm, which enables Navigator to detect most informative regions under the guidance from Teacher.
Ranked #42 on Fine-Grained Image Classification on FGVC Aircraft
no code implementations • 23 Mar 2017 • Yudong Liang, Ze Yang, Kai Zhang, Yihui He, Jinjun Wang, Nanning Zheng
To tackle with the second problem, a lightweight CNN architecture which has carefully designed width, depth and skip connections was proposed.