26 skills found
covid19datahub / COVID19A worldwide epidemiological database for COVID-19 at fine-grained spatial resolution
facebookresearch / VCMeshConvLearning latent representations of registered meshes is useful for many 3D tasks. Techniques have recently shifted to neural mesh autoencoders. Although they demonstrate higher precision than traditional methods, they remain unable to capture fine-grained deformations. Furthermore, these methods can only be applied to a template-specific surface mesh, and is not applicable to more general meshes, like tetrahedrons and non-manifold meshes. While more general graph convolution methods can be employed, they lack performance in reconstruction precision and require higher memory usage. In this paper, we propose a non-template-specific fully convolutional mesh autoencoder for arbitrary registered mesh data. It is enabled by our novel convolution and (un)pooling operators learned with globally shared weights and locally varying coefficients which can efficiently capture the spatially varying contents presented by irregular mesh connections. Our model outperforms state-of-the-art methods on reconstruction accuracy. In addition, the latent codes of our network are fully localized thanks to the fully convolutional structure, and thus have much higher interpolation capability than many traditional 3D mesh generation models.
appletea233 / LLaVA ST[CVPR 2025] LLaVA-ST: A Multimodal Large Language Model for Fine-Grained Spatial-Temporal Understanding
yuqian1023 / Deep SBIR TfImplementation of 'Sketch Me That Shoe' and 'Deep Spatial-Semantic Attention for Fine-grained Sketch-based Image Retrieval'
filipematias23 / CleanRfieldcleanRfield: This package is a compilation of functions to clean and filter observations from yield monitors or other agricultural spatial point data. Yield monitors are prone to error, and filtering the observations or removing observations from near field boundaries can improve estimates of whole-field yield, combine speed, grain moisture, or other parameters. In this package, users can easily select filters thresholding for one or more traits and prepare a smaller dataset to make decisions.
EasonXiao-888 / SpatialEditSpatialEdit: Benchmarking Fine-Grained Image Spatial Editing
VisionXLab / AirSpatialBot[TGRS'25] AirSpatialBot: A Spatially-Aware Aerial Agent for Fine-Grained Vehicle Attribute Recognization and Retrieval
BizhuWu / FineMotionFineMotion: A Dataset and Benchmark with both Spatial and Temporal Annotation for Fine-grained Motion Generation and Editing
iiasa / Ibis.iSDMModelling framework for creating (Integrated) SDMs
StatBiomed / FineSTFine-grained Spatial Transcriptomics by integrating paired histology image
CVI-SZU / FineMotionFineMotion: A Dataset and Benchmark with both Spatial and Temporal Annotation for Fine-grained Motion Generation and Editing
TauferLab / SOMOSPIESOMOSPIE (Soil Moisture Spatial Inference Engine) consists of a Jupyter Notebook and a suite of machine learning methods to process inputs of available coarse-grained soil moisture data at its native spatial resolution. Features include the selection of a geographic region of interest, prediction of missing values across the entire region of interest (i.e., gap-filling), analysis of generated fine-grained predictions, and visualization of both predictions and analyses.
reneschubert / KefluxA python package to compute the flux of oceanic kinetic energy across spatial scales from 2D velocity fields at a specific depth using a coarse-graining approach.
mustansarfiaz / PS ARMAbstract. Person search is a challenging problem with various real- world applications, that aims at joint person detection and re-identification of a query person from uncropped gallery images. Although, previous study focuses on rich feature information learning, it’s still hard to re- trieve the query person due to the occurrence of appearance deformations and background distractors. In this paper, we propose a novel attention- aware relation mixer (ARM) module for person search, which exploits the global relation between different local regions within RoI of a per- son and make it robust against various appearance deformations and occlusion. The proposed ARM is composed of a relation mixer block and a spatio-channel attention layer. The relation mixer block introduces a spatially attended spatial mixing and a channel-wise attended channel mixing for effectively capturing discriminative relation features within an RoI. These discriminative relation features are further enriched by intro- ducing a spatio-channel attention where the foreground and background discriminability is empowered in a joint spatio-channel space. Our ARM module is generic and it does not rely on fine-grained supervisions or topological assumptions, hence being easily integrated into any Faster R-CNN based person search methods. Comprehensive experiments are performed on two challenging benchmark datasets: CUHK-SYSU [1] and PRW [2]. Our PS-ARM achieves state-of-the-art performance on both datasets. On the challenging PRW dataset, our PS-ARM achieves an absolute gain of 5% in the mAP score over SeqNet, while operating at a comparable speed
Xovee / StcfTBDATA 2023. STCF: Spatial-Temporal Contrasting for Fine-Grained Urban Flow inference.
sjy-1995 / STAN OSFGR CodeThis is the code for our paper: "Spatial-Temporal Attention Network for Open-Set Fine-Grained Image Recognition", which is under review.
hedj17 / DSTThis project implements two dynamic spatiotemporal interpolation (DST) methods, i.e., coarse-grained DST (CGDST) and fine-grained DST (FGDST) using both temporal and spatial interpolation results. Different from other hybrid spatiotemporal interpolation methods, they make differences in the contribution of temporal and spatial interpolation results and assign them with different weights. Both CGDST and FGDST treat each missing value differently and fill it by considering the reliability of both temporal and spatial interpolation results in terms of the lengths of its column gap and row gap. CGDST treats each missing value in a continuous missing area equally and all missing values have same lengths of column and row gaps and FGDST goes beyond CGDST and treats each missing value differently based on its temporal distance to the nearest real observed values in both forward and backward directions.
shinying / Dest[BMVC 2022 Spotlight] Learning Fine-Grained Visual Understanding for Video Question Answering via Decoupling Spatial-Temporal Modeling
duongttr / SWP TF2UNOFFICIAL TensorFlow 2's implementation of Spatially Weighted Pooling (SWP) in paper "Deep CNNs With Spatially Weighted Pooling for Fine-Grained Car Recognition"
juliagviu / Finegrained Mat SelectionFine-Grained Spatially Varying Material Selection in Images
