More Like this

1. Self-supervised Feature Learning by Cross-modality and Cross-view Correspondences

   Jing, L; Zhang, L; Tian, Y (June 2021, The 4th Multimodal Learning and Applications (MULA) Workshop in conjunction with CVPR 2021)

   null (Ed.)

   The success of supervised learning requires large-scale ground truth labels which are very expensive, time- consuming, or may need special skills to annotate. To address this issue, many self- or un-supervised methods are developed. Unlike most existing self-supervised methods to learn only 2D image features or only 3D point cloud features, this paper presents a novel and effective self-supervised learning approach to jointly learn both 2D image features and 3D point cloud features by exploiting cross-modality and cross-view correspondences without using any human annotated labels. Specifically, 2D image features of rendered images from different views are extracted by a 2D convolutional neural network, and 3D point cloud features are extracted by a graph convolution neural network. Two types of features are fed into a two-layer fully connected neural network to estimate the cross-modality correspondence. The three networks are jointly trained (i.e. cross-modality) by verifying whether two sampled data of different modalities belong to the same object, meanwhile, the 2D convolutional neural network is additionally optimized through minimizing intra-object distance while maximizing inter-object distance of rendered images in different views (i.e. cross-view). The effectiveness of the learned 2D and 3D features is evaluated by transferring them on five different tasks including multi-view 2D shape recognition, 3D shape recognition, multi-view 2D shape retrieval, 3D shape retrieval, and 3D part-segmentation. Extensive evaluations on all the five different tasks across different datasets demonstrate strong generalization and effectiveness of the learned 2D and 3D features by the proposed self-supervised method. 

   more » « less
2. Fast Hand Detection in Collaborative Learning Environments

   https://doi.org/10.1007/978-3-030-89128-2_43  

   Teeparthi, S.; Jatla, V.; Pattichis, M.S.; Celedón-Pattichis, S.; Carlos LópezLeiva, C. (October 2021, CAIP 2021: Computer Analysis of Images and Patterns)

   Long-term object detection requires the integration of frame-based results over several seconds. For non-deformable objects, long-term detection is often addressed using object detection followed by video tracking. Unfortunately, tracking is inapplicable to objects that undergo dramatic changes in appearance from frame to frame. As a related example, we study hand detection over long video recordings in collaborative learning environments. More specifically, we develop long-term hand detection methods that can deal with partial occlusions and dramatic changes in appearance. Our approach integrates object-detection, followed by time projections, clustering, and small region removal to provide effective hand detection over long videos. The hand detector achieved average precision (AP) of 72% at 0.5 intersection over union (IoU). The detection results were improved to 81% by using our optimized approach for data augmentation. The method runs at 4.7× the real-time with AP of 81% at 0.5 intersection over the union. Our method reduced the number of false-positive hand detections by 80% by improving IoU ratios from 0.2 to 0.5. The overall hand detection system runs at 4× real-time. 

   more » « less
3. Automatic Segmentation of Sinkholes Using a Convolutional Neural Network

   https://doi.org/10.1029/2021EA002195  

   Rafique, Muhammad Usman; Zhu, Junfeng; Jacobs, Nathan (February 2022, Earth and Space Science)

   Abstract Sinkholes are the most abundant surface features in karst areas worldwide. Understanding sinkhole occurrences and characteristics is critical for studying karst aquifers and mitigating sinkhole‐related hazards. Most sinkholes appear on the land surface as depressions or cover collapses and are commonly mapped from elevation data, such as digital elevation models (DEMs). Existing methods for identifying sinkholes from DEMs often require two steps: locating surface depressions and separating sinkholes from non‐sinkhole depressions. In this study, we explored deep learning to directly identify sinkholes from DEM data and aerial imagery. A key contribution of our study is an evaluation of various ways of integrating these two types of raster data. We used an image segmentation model, U‐Net, to locate sinkholes. We trained separate U‐Net models based on four input images of elevation data: a DEM image, a slope image, a DEM gradient image, and a DEM‐shaded relief image. Three normalization techniques (Global, Gaussian, and Instance) were applied to improve the model performance. Model results suggest that deep learning is a viable method to identify sinkholes directly from the images of elevation data. In particular, DEM gradient data provided the best input for U‐net image segmentation models to locate sinkholes. The model using the DEM gradient image with Gaussian normalization achieved the best performance with a sinkhole intersection‐over‐union (IoU) of 45.38% on the unseen test set. Aerial images, however, were not useful in training deep learning models for sinkholes as the models using an aerial image as input achieved sinkhole IoUs below 3%. 

   more » « less
4. Enhanced U-Net Tool Segmentation using Hybrid Coordinate Representations of Endoscopic Images

   https://doi.org/10.1109/ISMR48346.2021.9661519  

   Huang, Kevin; Chitrakar, Digesh; Jiang, Wenfan; Su, Yun-Hsuan (November 2021, 2021 International Symposium on Medical Robotics (ISMR))

   This paper presents an approach to enhanced endoscopic tool segmentation combining separate pathways utilizing input images in two different coordinate representations. The proposed method examines U-Net convolutional neural networks with input endoscopic images represented via (1) the original rectangular coordinate format alongside (2) a morphological polar coordinate transformation. To maximize information and the breadth of the endoscope frustrum, imaging sensors are oftentimes larger than the image circle. This results in unused border regions. Ideally, the region of interest is proximal to the image center. The above two observations formed the basis for the morphological polar transformation pathway as an augmentation to typical rectangular input image representations. Results indicate that neither of the two investigated coordinate representations consistently yielded better segmentation performance as compared to the other. Improved segmentation can be achieved with a hybrid approach that carefully selects which of the two pathways to be used for individual input images. Towards that end, two binary classifiers were trained to identify, given an input endoscopic image, which of the two coordinate representation segmentation pathways (rectangular or polar), would result in better segmentation performance. Results are promising and suggest marked improvements using a hybrid pathway selection approach compared to either alone. The experiment used to evaluate the proposed hybrid method utilized a dataset consisting of 8360 endoscopic images from real surgery and evaluated segmentation performance with Dice coefficient and Intersection over Union. The results suggest that on-the-fly polar transformation for tool segmentation is useful when paired with the proposed hybrid tool-segmentation approach. 

   more » « less
5. Facilitating cell segmentation with the projection-enhancement network

   https://doi.org/10.1088/1478-3975/acfe53  

   Eddy, Christopher Z; Naylor, Austin; Cunningham, Christian T; Sun, Bo (October 2023, Physical Biology)

   Abstract Contemporary approaches to instance segmentation in cell science use 2D or 3D convolutional networks depending on the experiment and data structures. However, limitations in microscopy systems or efforts to prevent phototoxicity commonly require recording sub-optimally sampled data that greatly reduces the utility of such 3D data, especially in crowded sample space with significant axial overlap between objects. In such regimes, 2D segmentations are both more reliable for cell morphology and easier to annotate. In this work, we propose the projection enhancement network (PEN), a novel convolutional module which processes the sub-sampled 3D data and produces a 2D RGB semantic compression, and is trained in conjunction with an instance segmentation network of choice to produce 2D segmentations. Our approach combines augmentation to increase cell density using a low-density cell image dataset to train PEN, and curated datasets to evaluate PEN. We show that with PEN, the learned semantic representation in CellPose encodes depth and greatly improves segmentation performance in comparison to maximum intensity projection images as input, but does not similarly aid segmentation in region-based networks like Mask-RCNN. Finally, we dissect the segmentation strength against cell density of PEN with CellPose on disseminated cells from side-by-side spheroids. We present PEN as a data-driven solution to form compressed representations of 3D data that improve 2D segmentations from instance segmentation networks. 

   more » « less