activity
20172022
most citedWhy Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognition

99 citations · 290 across the 12 of their papers we have counts for

collaborators

24 papers

cs.AI202280 cited

ESCM: Entire Space Counterfactual Multi-Task Model for Post-Click Conversion Rate Estimation

Hao Wang, Tai-Wei Chang, Tianqiao Liu +5

Accurate estimation of post-click conversion rate is critical for building recommender systems, which has long been confronted with sample selection bias and data sparsity issues.…

cs.CV20202 cited

Instance-aware Image Colorization

Jheng-Wei Su, Hung-Kuo Chu, Jia-Bin Huang

Image colorization is inherently an ill-posed problem with multi-modal uncertainty. Previous methods leverage the deep neural network to map input grayscale images to plausible col…

cs.CV2020

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…

cs.CV2020

CrDoCo: Pixel-level Domain Transfer with Cross-Domain Consistency

Yun-Chun Chen, Yen-Yu Lin, Ming-Hsuan Yang +1

Unsupervised domain adaptation algorithms aim to transfer the knowledge learned from one domain to another (e.g., synthetic to real images). The adapted representations often do no…

cs.CV2020

Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformation

Hung-Yu Tseng, Hsin-Ying Lee, Jia-Bin Huang +1

Few-shot classification aims to recognize novel categories with only few labeled images in each class. Existing metric-based few-shot classification algorithms predict categories b…

cs.CV201999 cited

Why Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognition

Jinwoo Choi, Chen Gao, Joseph C. E. Messou +1

Human activities often occur in specific scene contexts, e.g., playing basketball on a basketball court. Training a model using existing video datasets thus inevitably captures and…