3.5k citations · 4.8k across the 85 of their papers we have counts for
35 papers · 1 filter
Online Adaptation for Consistent Mesh Reconstruction in the Wild
Xueting Li, Sifei Liu, Shalini De Mello +4
This paper presents an algorithm to reconstruct temporally consistent 3D meshes of deformable object instances from videos in the wild. Without requiring annotations of 3D mesh, 2D…
D2-Net: Weakly-Supervised Action Localization via Discriminative Embeddings and Denoised Activations
Sanath Narayan, Hisham Cholakkal, Munawar Hayat +3
This work proposes a weakly-supervised temporal action localization framework, called D2-Net, which strives to temporally localize actions using video-level supervision. Our main c…
Unsupervised Discovery of Disentangled Manifolds in GANs
Yu-Ding Lu, Hsin-Ying Lee, Hung-Yu Tseng +1
As recent generative models can generate photo-realistic images, people seek to understand the mechanism behind the generation process. Interpretable generation process is benefici…
Shaping Deep Feature Space towards Gaussian Mixture for Visual Classification
Weitao Wan, Jiansheng Chen, Cheng Yu +3
The softmax cross-entropy loss function has been widely used to train deep models for various tasks. In this work, we propose a Gaussian mixture (GM) loss function for deep neural…
Continuous and Diverse Image-to-Image Translation via Signed Attribute Vectors
Qi Mao, Hung-Yu Tseng, Hsin-Ying Lee +3
Recent image-to-image (I2I) translation algorithms focus on learning the mapping from a source to a target domain. However, the continuous translation problem that synthesizes inte…
Unsupervised Domain Adaptation for Spatio-Temporal Action Localization
Nakul Agarwal, Yi-Ting Chen, Behzad Dariush +1
Spatio-temporal action localization is an important problem in computer vision that involves detecting where and when activities occur, and therefore requires modeling of both spat…
Multi-path Neural Networks for On-device Multi-domain Visual Classification
Qifei Wang, Junjie Ke, Joshua Greaves +9
Learning multiple domains/tasks with a single model is important for improving data efficiency and lowering inference cost for numerous vision tasks, especially on resource-constra…
Every Pixel Matters: Center-aware Feature Alignment for Domain Adaptive Object Detector
Cheng-Chun Hsu, Yi-Hsuan Tsai, Yen-Yu Lin +1
A domain adaptive object detector aims to adapt itself to unseen domains that may contain variations of object appearance, viewpoints or backgrounds. Most existing methods adopt fe…
SoDA: Multi-Object Tracking with Soft Data Association
Wei-Chih Hung, Henrik Kretzschmar, Tsung-Yi Lin +4
Robust multi-object tracking (MOT) is a prerequisite fora safe deployment of self-driving cars. Tracking objects, however, remains a highly challenging problem, especially in clutt…
Learning to Caricature via Semantic Shape Transform
Wenqing Chu, Wei-Chih Hung, Yi-Hsuan Tsai +4
Caricature is an artistic drawing created to abstract or exaggerate facial features of a person. Rendering visually pleasing caricatures is a difficult task that requires professio…
Mixup-CAM: Weakly-supervised Semantic Segmentation via Uncertainty Regularization
Yu-Ting Chang, Qiaosong Wang, Wei-Chih Hung +3
Obtaining object response maps is one important step to achieve weakly-supervised semantic segmentation using image-level labels. However, existing methods rely on the classificati…
Weakly-Supervised Semantic Segmentation via Sub-category Exploration
Yu-Ting Chang, Qiaosong Wang, Wei-Chih Hung +3
Existing weakly-supervised semantic segmentation methods using image-level annotations typically rely on initial responses to locate object regions. However, such response maps gen…
Spatiotemporal Contrastive Video Representation Learning
Rui Qian, Tianjian Meng, Boqing Gong +4
We present a self-supervised Contrastive Video Representation Learning (CVRL) method to learn spatiotemporal visual representations from unlabeled videos. Our representations are l…
Learnable Cost Volume Using the Cayley Representation
Taihong Xiao, Jinwei Yuan, Deqing Sun +4
Cost volume is an essential component of recent deep models for optical flow estimation and is usually constructed by calculating the inner product between two feature vectors. How…
Controllable Image Synthesis via SegVAE
Yen-Chi Cheng, Hsin-Ying Lee, Min Sun +1
Flexible user controls are desirable for content creation and image editing. A semantic map is commonly used intermediate representation for conditional image generation. Compared…
RetrieveGAN: Image Synthesis via Differentiable Patch Retrieval
Hung-Yu Tseng, Hsin-Ying Lee, Lu Jiang +2
Image generation from scene description is a cornerstone technique for the controlled generation, which is beneficial to applications such as content creation and image editing. In…
Modeling Artistic Workflows for Image Generation and Editing
Hung-Yu Tseng, Matthew Fisher, Jingwan Lu +3
People often create art by following an artistic workflow involving multiple stages that inform the overall design. If an artist wishes to modify an earlier decision, significant w…
Semi-Supervised Learning with Meta-Gradient
Xin-Yu Zhang, Taihong Xiao, Haolin Jia +2
In this work, we propose a simple yet effective meta-learning algorithm in semi-supervised learning. We notice that most existing consistency-based approaches suffer from overfitti…
WW-Nets: Dual Neural Networks for Object Detection
Mohammad K. Ebrahimpour, J. Ben Falandays, Samuel Spevack +2
We propose a new deep convolutional neural network framework that uses object location knowledge implicit in network connection weights to guide selective attention in object detec…
Multi-Scale Boosted Dehazing Network with Dense Feature Fusion
Hang Dong, Jinshan Pan, Lei Xiang +4
In this paper, we propose a Multi-Scale Boosted Dehazing Network with Dense Feature Fusion based on the U-Net architecture. The proposed method is designed based on two principles,…
Regularizing Meta-Learning via Gradient Dropout
Hung-Yu Tseng, Yi-Wen Chen, Yi-Hsuan Tsai +3
With the growing attention on learning-to-learn new tasks using only a few examples, meta-learning has been widely used in numerous problems such as few-shot classification, reinfo…
Learning to See Through Obstructions
Yu-Lun Liu, Wei-Sheng Lai, Ming-Hsuan Yang +2
We present a learning-based approach for removing unwanted obstructions, such as window reflections, fence occlusions or raindrops, from a short sequence of images captured by a mo…
Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipeline
Yu-Lun Liu, Wei-Sheng Lai, Yu-Sheng Chen +4
Recovering a high dynamic range (HDR) image from a single low dynamic range (LDR) input image is challenging due to missing details in under-/over-exposed regions caused by quantiz…
Deep Semantic Matching with Foreground Detection and Cycle-Consistency
Yun-Chun Chen, Po-Hsiang Huang, Li-Yu Yu +3
Establishing dense semantic correspondences between object instances remains a challenging problem due to background clutter, significant scale and pose differences, and large intr…