We propose DSP-SLAM, an object-oriented SLAM system that builds a rich and accurate joint map of dense 3D models for foreground objects, and sparse landmark points to represent the background. DSP-SLAM takes as input the 3D point cloud reconstructed
by a feature-based SLAM system and equips it with the ability to enhance its sparse map with dense reconstructions of detected objects. Objects are detected via semantic instance segmentation, and their shape and pose is estimated using category-specific deep shape embeddings as priors, via a novel second order optimization. Our object-aware bundle adjustment builds a pose-graph to jointly optimize camera poses, object locations and feature points. DSP-SLAM can operate at 10 frames per second on 3 different input modalities: monocular, stereo, or stereo+LiDAR. We demonstrate DSP-SLAM operating at almost frame rate on monocular-RGB sequences from the Friburg and Redwood-OS datasets, and on stereo+LiDAR sequences on the KITTI odometry dataset showing that it achieves high-quality full object reconstructions, even from partial observations, while maintaining a consistent global map. Our evaluation shows improvements in object pose and shape reconstruction with respect to recent deep prior-based reconstruction methods and reductions in camera tracking drift on the KITTI dataset.
Cancer diagnosis, prognosis, and therapeutic response predictions are based on morphological information from histology slides and molecular profiles from genomic data. However, most deep learning-based objective outcome prediction and grading paradi
gms are based on histology or genomics alone and do not make use of the complementary information in an intuitive manner. In this work, we propose Pathomic Fusion, an interpretable strategy for end-to-end multimodal fusion of histology image and genomic (mutations, CNV, RNA-Seq) features for survival outcome prediction. Our approach models pairwise feature interactions across modalities by taking the Kronecker product of unimodal feature representations and controls the expressiveness of each representation via a gating-based attention mechanism. Following supervised learning, we are able to interpret and saliently localize features across each modality, and understand how feature importance shifts when conditioning on multimodal input. We validate our approach using glioma and clear cell renal cell carcinoma datasets from the Cancer Genome Atlas (TCGA), which contains paired whole-slide image, genotype, and transcriptome data with ground truth survival and histologic grade labels. In a 15-fold cross-validation, our results demonstrate that the proposed multimodal fusion paradigm improves prognostic determinations from ground truth grading and molecular subtyping, as well as unimodal deep networks trained on histology and genomic data alone. The proposed method establishes insight and theory on how to train deep networks on multimodal biomedical data in an intuitive manner, which will be useful for other problems in medicine that seek to combine heterogeneous data streams for understanding diseases and predicting response and resistance to treatment.
Temporal sentence grounding in videos aims to detect and localize one target video segment, which semantically corresponds to a given sentence. Existing methods mainly tackle this task via matching and aligning semantics between a sentence and candid
ate video segments, while neglect the fact that the sentence information plays an important role in temporally correlating and composing the described contents in videos. In this paper, we propose a novel semantic conditioned dynamic modulation (SCDM) mechanism, which relies on the sentence semantics to modulate the temporal convolution operations for better correlating and composing the sentence related video contents over time. More importantly, the proposed SCDM performs dynamically with respect to the diverse video contents so as to establish a more precise matching relationship between sentence and video, thereby improving the temporal grounding accuracy. Extensive experiments on three public datasets demonstrate that our proposed model outperforms the state-of-the-arts with clear margins, illustrating the ability of SCDM to better associate and localize relevant video contents for temporal sentence grounding. Our code for this paper is available at https://github.com/yytzsy/SCDM .
How to efficiently generate an accurate, well-structured overview report (ORPT) over thousands of related documents is challenging. A well-structured ORPT consists of sections of multiple levels (e.g., sections and subsections). None of the existing
multi-document summarization (MDS) algorithms is directed toward this task. To overcome this obstacle, we present NDORGS (Numerous Documents Overview Report Generation Scheme) that integrates text filtering, keyword scoring, single-document summarization (SDS), topic modeling, MDS, and title generation to generate a coherent, well-structured ORPT. We then devise a multi-criteria evaluation method using techniques of text mining and multi-attribute decision making on a combination of human judgments, running time, information coverage, and topic diversity. We evaluate ORPTs generated by NDORGS on two large corpora of documents, where one is classified and the other unclassified. We show that, using Saatys pairwise comparison 9-point scale and under TOPSIS, the ORPTs generated on SDSs with the length of 20% of the original documents are the best overall on both datasets.
This paper describes our solution for the 2$^text{nd}$ YouTube-8M video understanding challenge organized by Google AI. Unlike the video recognition benchmarks, such as Kinetics and Moments, the YouTube-8M challenge provides pre-extracted visual and
audio features instead of raw videos. In this challenge, the submitted model is restricted to 1GB, which encourages participants focus on constructing one powerful single model rather than incorporating of the results from a bunch of models. Our system fuses six different sub-models into one single computational graph, which are categorized into three families. More specifically, the most effective family is the model with non-local operations following the NetVLAD encoding. The other two family models are Soft-BoF and GRU, respectively. In order to further boost single models performance, the model parameters of different checkpoints are averaged. Experimental results demonstrate that our proposed system can effectively perform the video classification task, achieving 0.88763 on the public test set and 0.88704 on the private set in terms of GAP@20, respectively. We finally ranked at the fourth place in the YouTube-8M video understanding challenge.
