• Publications
  • Influence
Domain Adaptive Faster R-CNN for Object Detection in the Wild
TLDR
This work builds on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy, based on $$-divergence theory.
Semantic Foggy Scene Understanding with Synthetic Data
TLDR
A complete pipeline to add synthetic fog to real, clear-weather images using incomplete depth information is developed and supervised learning with the authors' synthetic data significantly improves the performance of state-of-the-art CNN for SFSU on Foggy Driving.
Fast Optical Flow Using Dense Inverse Search
TLDR
The Dense Inverse Search-based method (DIS) is the efficient search of correspondences inspired by the inverse compositional image alignment proposed by Baker and Matthews (2001, 2004), making DIS ideal for real-time applications.
Jointly Optimized Regressors for Image Super‐resolution
TLDR
A collection of regressors are jointly learned, which collectively yield the smallest super‐resolving error for all training data, and this method is conceptually simple and computationally efficient, yet very effective.
Satellite Image Classification via Two-Layer Sparse Coding With Biased Image Representation
TLDR
This letter presents a method for satellites image classification involving visual attention into the satellite image classification; biologically inspired saliency information is exploited in the phase of the image representation, making the method more concentrated on the interesting objects and structures.
Ensemble Projection for Semi-supervised Image Classification
  • Dengxin Dai, L. Gool
  • Computer Science
    IEEE International Conference on Computer Vision
  • 1 December 2013
TLDR
This method consistently outperforms previous methods for semi-supervised image classification, and lets itself combine well with these methods, and works well for self-taught image classification where unlabeled data are not coming from the same distribution as labeled ones, but rather from a random collection of images.
Unified Hypersphere Embedding for Speaker Recognition
TLDR
Results of experiments suggest that simple repetition and random time-reversion of utterances can reduce prediction errors by up to 18% and proposed logistic margin loss function leads to unified embeddings with state-of-the-art identification and competitive verification accuracies.
End-to-End Learning of Driving Models with Surround-View Cameras and Route Planners
TLDR
360-degree surround-view cameras help avoid failures made with a single front-view camera, in particular for city driving and intersection scenarios; and route planners help the driving task significantly, especially for steering angle prediction.
Dark Model Adaptation: Semantic Image Segmentation from Daytime to Nighttime
  • Dengxin Dai, L. Gool
  • Computer Science, Environmental Science
    21st International Conference on Intelligent…
  • 5 October 2018
TLDR
A novel method to progressive adapt the semantic models trained on daytime scenes, along with large-scale annotations therein, to nighttime scenes via the bridge of twilight time, to alleviate the cost of human annotation for nighttime images by transferring knowledge from standard daytime conditions.
Guided Curriculum Model Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation
TLDR
This work addresses the problem of semantic segmentation of nighttime images and improve the state-of-the-art, by adapting daytime models to nighttime without using nighttime annotations, and designs a new evaluation framework to address the substantial uncertainty of semantics in nighttime images.
...
1
2
3
4
5
...