Michael Opitz

We introduce a simple yet effective fusion method of LiDAR and RGB data to segment LiDAR point clouds. Utilizing the dense native range representation of a LiDAR sensor and the setup calibration, we establish point correspondences between the two input modalities. Subsequently, we are able to warp and fuse the features from one domain into the other. Therefore, we can jointly exploit information from both data sources within one single network. To show the merit of our method, we extend SqueezeSeg, a point cloud segmentation network, with an RGB feature branch and fuse it into the original structure. Our extension called FuseSeg leads to an improvement of up to 18% IoU on the KITTI benchmark. In addition to the improved accuracy, we also achieve real-time performance at 50 fps, five times as fast as the KITTI LiDAR data recording speed.

Learning similarity functions between image pairs with deep neural networks yields highly correlated activations of embeddings. In this work, we show how to improve the robustness of such embeddings by exploiting the independence within ensembles. To this end, we divide the last embedding layer of a deep network into an embedding ensemble and formulate the task of training this ensemble as an online gradient boosting problem. Each learner receives a reweighted training sample from the previous learners. Further, we propose two loss functions which increase the diversity in our ensemble. These loss functions can be applied either for weight initialization or during training. Together, our contributions leverage large embedding sizes more effectively by significantly reducing correlation of the embedding and consequently increase retrieval accuracy of the embedding. Our method works with any differentiable loss function and does not introduce any additional parameters during test time. We evaluate our metric learning method on image retrieval tasks and show that it improves over state-of-the-art methods on the CUB-200-2011, Cars-196, Stanford Online Products, In-Shop Clothes Retrieval and VehicleID datasets. Therefore, our findings suggest that by dividing deep networks at the end into several smaller and diverse networks, we can significantly reduce overfitting.

Learning similarity functions between image pairs with deep neural networks yields highly correlated activations of embeddings. In this work, we show how to improve the robustness of such embeddings by exploiting the independence within ensembles. To this end, we divide the last embedding layer of a deep network into an embedding ensemble and formulate training this ensemble as an online gradient boosting problem. Each learner receives a reweighted training sample from the previous learners. Further, we propose two loss functions which increase the diversity in our ensemble. These loss functions can be applied either for weight initialization or during training. Together, our contributions leverage large embedding sizes more effectively by significantly reducing correlation of the embedding and consequently increase retrieval accuracy of the embedding. Our method works with any differentiable loss function and does not introduce any additional parameters during test time. We evaluate our metric learning method on image retrieval tasks and show that it improves over state-of-the-art methods on the CUB 200-2011, Cars-196, Stanford Online Products, In-Shop Clothes Retrieval and VehicleID datasets.

Learning similarity functions between image pairs with deep neural networks yields highly correlated activations of large embeddings. In this work, we show how to improve the robustness of embeddings by exploiting independence in ensembles. We divide the last embedding layer of a deep network into an embedding ensemble and formulate training this ensemble as an online gradient boosting problem. Each learner receives a reweighted training sample from the previous learners. This leverages large embedding sizes more effectively by significantly reducing correlation of the embedding and consequently increases retrieval accuracy of the embedding. Our method does not introduce any additional parameters and works with any differentiable loss function. We evaluate our metric learning method on image retrieval tasks and show that it improves over state-ofthe- art methods on the CUB-200-2011, Cars-196, Stanford Online Products, In-Shop Clothes Retrieval and VehicleID datasets by a significant margin.

Detection of partially occluded objects is a challenging computer vision problem. Standard Convolutional Neural Network (CNN) detectors fail if parts of the detection window are occluded, since not every sub-part of the window is discriminative on its own. To address this issue, we propose a novel loss layer for CNNs, named grid loss, which minimizes the error rate on sub-blocks of a convolution layer independently rather than over the whole feature map. This results in parts being more discriminative on their own, enabling the detector to recover if the detection window is partially occluded. By mapping our loss layer back to a regular fully connected layer, no additional computational cost is incurred at runtime compared to standard CNNs. We demonstrate our method for face detection on several public face detection benchmarks and show that our method outperforms regular CNNs, is suitable for realtime applications and achieves state-of-the-art performance.

Text recognition in natural scene images is an application for several computer vision applications like licence plate recognition, automated translation of street signs, help for visually impaired people or image retrieval. In this work an end-to-end text recognition system is presented. For detection an AdaBoost ensemble with a modified Local Ternary Pattern (LTP) feature-set with a post-processing stage build upon Maximally Stable Extremely Region (MSER) is used. The text recognition is done using a deep Convolution Neural Network (CNN) trained with backpropagation. The system presented outperforms state of the art methods on the ICDAR 2003 dataset in the text-detection (F-Score: 74.2%), dictionary-driven cropped-word recognition (F-Score: 87.1%) and dictionary-driven end-to-end recognition (F-Score: 72.6%) tasks.