We present PartNet: a consistent, large-scale dataset of 3D objects annotated with fine-grained, instance-level, and hierarchical 3D part information. Our dataset consists of 573,585 part instances over 26,671 3D models covering 24 object categories. This dataset enables and serves as a catalyst for many tasks such as shape analysis, dynamic 3D scene modeling and simulation, affordance analysis, and others. Using our dataset, we establish three benchmarking tasks for evaluating 3D part recognition: fine-grained semantic segmentation, hierarchical semantic segmentation, and instance segmentation. We benchmark four state-ofthe-art 3D deep learning algorithms for fine-grained semantic segmentation and three baseline methods for hierarchical semantic segmentation. We also propose a novel method for part instance segmentation and demonstrate its superior performance over existing methods.

Generative adversarial networks (GANs) transform latent vectors into visually plausible images. It is generally thought that the original GAN formulation gives no out-of-the-box method to reverse the mapping, projecting images back into latent space. We introduce a simple, gradient-based technique called stochastic clipping. In experiments, for images generated by the GAN, we precisely recover their latent vector pre-images 100% of the time. Additional experiments demonstrate that this method is robust to noise. Finally, we show that even for unseen images, our method appears to recover unique encodings.

Sign language recognition is important for natural and convenient communication between deaf community and hearing majority. We take the highly efficient initial step of automatic fingerspelling recognition system using convolutional neural networks (CNNs) from depth maps. In this work, we consider relatively larger number of classes compared with the previous literature. We train CNNs for the classification of 31 alphabets and numbers using a subset of collected depth data from multiple subjects. While using different learning configurations, such as hyper-parameter selection with and without validation, we achieve 99.99% accuracy for observed signers and 83.58% to 85.49% accuracy for new signers. The result shows that accuracy improves as we include more data from different subjects during training. The processing time is 3 ms for the prediction of a single image. To the best of our knowledge, the system achieves the highest accuracy and speed. The trained model and dataset is available on our repository 1 .

No abstract

Deep Convolutional Neural Networks (CNN) are the state-of-the-art performers for the object detection task. It is well known that object detection requires more computation and memory than image classification. In this work, we propose LCDet, a fully-convolutional neural network for generic object detection that aims to work in embedded systems. We design and develop an end-to-end TensorFlow(TF)-based model. The detection works by a single forward pass through the network. Additionally, we employ 8-bit quantization on the learned weights. As a use case, we choose face detection and train the proposed model on images containing a varying number of faces of different sizes. We evaluate the face detection performance on publicly available dataset FDDB and Widerface. Our experimental results show that the proposed method achieves comparative accuracy comparing with state-ofthe-art CNN-based face detection methods while reducing the model size by 3× and memory-BW by 3 − 4× comparing with one of the best real-time CNN-based object detector YOLO [23]. Our 8-bit fixed-point TF-model provides additional 4× memory reduction while keeping the accuracy nearly as good as the floating point model and achieves 20× performance gain compared to the floating point model. Thus the proposed model is amenable for embedded implementations and is generic to be extended to any number of categories of objects.