Find research datasets worth reusing

Figure 3 in Use of open-source object detection algorithms to detect Palmer amaranth (Amoronthus polmeri) in soybean

Figure 3. Intersection over union (IoU) equation, defined as the overlap between the ground truth annotation and the computer prediction bounding box, divided by the total area of the two bounding boxes.IoU overlaps greater than 0.5 were considered true-positive predictions,whereas overlaps less than 0.5 were considered false-positive predictions.

Figure 2 in Use of open-source object detection algorithms to detect Palmer amaranth (Amoronthus polmeri) in soybean

Figure 2. Illustration of the annotation process. Amoronthus polmeri and soybean plants are labeled in this figure with orange and white boxes, respectively. Bounding boxes overlap with neighboring bounding boxes when plant features are irregular. In cases where a single bounding box could not encompass a plant without including a plant of another species, multiple irregular bounding boxes were drawn on a single specimen.

Figure 4 in Use of open-source object detection algorithms to detect Palmer amaranth (Amoronthus polmeri) in soybean

Figure 4. Mean average precision (mAP) results of each model after training. YOLOv5 was considered the best-performing algorithm of each tested model with a mAP of 0.77.

Figure 6. Precision–recall curve for YOLOv5. Amoronthus polmeri achieved a in Use of open-source object detection algorithms to detect Palmer amaranth (Amoronthus polmeri) in soybean

Figure 6. Precision–recall curve for YOLOv5. Amoronthus polmeri achieved a slightly higher average precision (AP) (0.788) than soybean. Solid blue line represents mean average precision (mAP) computed on the test data set. The AP for each class and the mAP for the overall algorithm were representative of the area of the graph under each respective curve.

Figure 5 in Use of open-source object detection algorithms to detect Palmer amaranth (Amoronthus polmeri) in soybean

Figure 5. Change in mean average precision (mAP) @ 0.5 over each epoch during training. mAP was reported after the completion of each epoch. Training was terminated after visual inspection of curve and when mAP @ 0.5 curve was seen to "plateau."

opencc-zeroJun 2024View details →

dryad40/100

Data from: SimPLE: A visuotactile method learned in simulation to precisely pick, localize, regrasp, and place objects

Existing robotic systems have a clear tension between generality and precision. Deployed solutions for robotic manipulation tend to fall into the paradigm of one robot solving a single task, lacking precise generalization, i.e., the ability to solve many tasks without compromising on precision. This paper explores solutions for precise and general pick-and-place. In precise pick-and-place, i.e. kitting, the robot transforms an unstructured arrangement of objects into an organized arrangement, which can facilitate further manipulation. We propose simPLE (simulation to Pick Localize and PLacE) as a solution to precise pick-and-place. simPLE learns to pick, regrasp and place objects precisely, given only the object CAD model and no prior experience. We develop three main components: task-aware grasping, visuotactile perception, and regrasp planning. Task-aware grasping computes affordances of grasps that are stable, observable, and favorable to placing. The visuotactile perception model relies on matching real observations against a set of simulated ones through supervised learning. Finally, we compute the desired robot motion by solving a shortest path problem on a graph of hand-to-hand regrasps. On a dual-arm robot equipped with visuotactile sensing, we demonstrate pick-and-place of 15 diverse objects with simPLE. The objects span a wide range of shapes and simPLE achieves successful placements into structured arrangements with 1mm clearance over 90% of the time for 6 objects, and over 80% of the time for 11 objects.

opencc-by-4.0Jun 2024View details →

The ORBIT (Object Recognition for Blind Image Training)-India Dataset

<div> The ORBIT (Object Recognition for Blind Image Training) -India Dataset is a collection of 105,243 images of 76 commonly used objects, collected by 12 individuals in India who are blind or have low vision. This dataset is an "Indian subset" of the original ORBIT dataset [1, 2], which was collected in the UK and Canada. In contrast to the ORBIT dataset, which was created in a Global North, Western, and English-speaking context, the ORBIT-India dataset features images taken in a low-resource, non-English-speaking, Global South context, a home to 90% of the world’s population of people with blindness. Since it is easier for blind or low-vision individuals to gather high-quality data by recording videos, this dataset, like the ORBIT dataset, contains images (each sized 224x224) derived from 587 videos. These videos were taken by our data collectors from various parts of India using the Find My Things [3] Android app. Each data collector was asked to record eight videos of at least 10 objects of their choice.  </div> <div> Collected between July and November 2023, this dataset represents a set of objects commonly used by people who are blind or have low vision in India, including earphones, talking watches, toothbrushes, and typical Indian household items like a belan (rolling pin), and a steel glass. These videos were taken in various settings of the data collectors' homes and workspaces using the Find My Things Android app.  </div> <div> The image dataset is stored in the ‘Dataset’ folder, organized by folders assigned to each data collector (P1, P2, ...P12) who collected them. Each collector's folder includes sub-folders named with the object labels as provided by our data collectors. Within each object folder, there are two subfolders: ‘clean’ for images taken on clean surfaces and ‘clutter’ for images taken in cluttered environments where the objects are typically found. The annotations are saved inside a  ‘Annotations’ folder containing a JSON file per video (e.g., P1--coffee mug--clean--231220_084852_coffee mug_224.json) that contains keys corresponding to all frames/images in that video (e.g., "P1--coffee mug--clean--231220_084852_coffee mug_224--000001.jpeg": {"object_not_present_issue": false, "pii_present_issue": false}, "P1--coffee mug--clean--231220_084852_coffee mug_224--000002.jpeg": {"object_not_present_issue": false, "pii_present_issue": false}, ...). The ‘object_not_present_issue’ key is True if the object is not present in the image, and the ‘pii_present_issue’ key is True, if there is a personally identifiable information (PII) present in the image. Note, all PII present in the images has been blurred to protect the identity and privacy of our data collectors. This dataset version was created by cropping images originally sized at 1080 × 1920; therefore, an unscaled version of the dataset will follow soon.  </div> <div> This project was funded by the Engineering and Physical Sciences Research Council (EPSRC) Industrial ICASE Award with Microsoft Research UK Ltd. as the Industrial Project Partner. We would like to acknowledge and express our gratitude to our data collectors for their efforts and time invested in carefully collecting videos to build this dataset for their community. The dataset is designed for developing few-shot learning algorithms, aiming to support researchers and developers in advancing object-recognition systems. We are excited to share this dataset and would love to hear from you if and how you use this dataset. Please feel free to reach out if you have any questions, comments or suggestions.  </div> <div> REFERENCES:  </div> <div> <ol> <li> Daniela Massiceti, Lida Theodorou, Luisa Zintgraf, Matthew Tobias Harris, Simone Stumpf, Cecily Morrison, Edward Cutrell, and Katja Hofmann. 2021. ORBIT: A real-world few-shot dataset for teachable object recognition collected from people who are blind or low vision. DOI: <a href="https://doi.org/10.25383/city.14294597" target="_blank" rel="noreferrer noopener">https://doi.org/10.25383/city.14294597</a> </li> <li> microsoft/ORBIT-Dataset. <a href="https://github.com/microsoft/ORBIT-Dataset" target="_blank" rel="noreferrer noopener">https://github.com/microsoft/ORBIT-Dataset</a>   </li> <li> Linda Yilin Wen, Cecily Morrison, Martin Grayson, Rita Faia Marques, Daniela Massiceti, Camilla Longden, and Edward Cutrell. 2024. Find My Things: Personalized Accessibility through Teachable AI for People who are Blind or Low Vision. In Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems (CHI EA '24). Association for Computing Machinery, New York, NY, USA, Article 403, 1–6. <a href="https://doi.org/10.1145/3613905.3648641" target="_blank" rel="noreferrer noopener">https://doi.org/10.1145/3613905.3648641</a>  </li> </ol> </div>

opencc-by-4.0Dec 2019View details →

Fig.6 in Reaching The Climate Objectives Of Forest Management In Latvia Within The Scope Of European Climate Policy

Fig.6. CO2 emissions in Latvia by sources in 2016, CO2 kt eq (Source: Created by authors after National GHG inventories).

opencc-by-4.0Dec 2019View details →

Fig.2 in Reaching The Climate Objectives Of Forest Management In Latvia Within The Scope Of European Climate Policy

Fig.2. Scheme of LFSRI Silava changes in forest resources projections process based on Nation Forest Inventory data (Source: LFSRI Silava).

opencc-by-4.0Dec 2019View details →

Fig.1 in Reaching The Climate Objectives Of Forest Management In Latvia Within The Scope Of European Climate Policy

Fig.1. Policy documents affecting the Climate Change Mitigation policy of the Forest sector by 2020.

opencc-by-4.0Dec 2010View details →

Figure 4 in Dactylorhiza Fuchsii (Druce) Soó As A Model Object In In Vitro Culture Study For Development Of Terrestrial Orchids

Figure 4. Root of D. fuchsii growing in the Figure 5. Root of D. fuchsii one year after exposition in NBG: arrow indicates elaborate transplanting ex vitro: arrow indicates some coiled structures known as pelotons of peloton-like structure as they do within the orchid mycorrhizal fungi in the root cortex (section was cells (section was stained with Trypan Blue; bar stained with Trypan Blue; bar = 50 mm) = 50 mm).

opencc-by-4.0Dec 2010View details →

Figure 1 in Dactylorhiza Fuchsii (Druce) Soó As A Model Object In In Vitro Culture Study For Development Of Terrestrial Orchids

Figure 1. Development of Dactylorhiza fuchsii (Druce) Soó in vitro conditions: A – beginning of morphogenesis after germination with development of protocorms, characteristic for orchids; B – initialization of rooting; C – sterile plantlets after exposition at 2 °C in the dark, simulating the natural dormancy period; D – plantlet before transplanting ex vitro (ro – root; tu – tuber).

opencc-by-4.0Dec 2020View details →

Figure 2 in Objective record of Epinephelus marginatus (Serranidae: Epinephelinae) from the Sultanate of Oman (Arabian Sea)

Figure 2. – Specimens of dusky grouper from Salalah, Oman. A: Epinephelus marginatus, 793 mm TL (MNHN-ICOS-01381). B: E. marginatus, 331 mm TL (MNHN- IC-2019-0025). C: Three individuals of E. marginatus observed at the Salalah fish market on 19/01/2019.

opencc-by-4.0Jul 2024View details →

Transfer learning with generative models for object detection on limited datasets

The provided datasets are used for the analysis in the work "Transfer learning with generative models for object detection on limited datasets" (https://doi.org/10.1088/2632-2153/ad65b5). The availability of data is limited in some fields, especially for object detection tasks, where it is necessary to have correctly labeled bounding boxes around each object. A notable example of such data scarcity is found in the domain of marine biology, where it is useful to develop methods to automatically detect submarine species for environmental monitoring. To address this data limitation, the state-of-the-art machine learning strategies employ two main approaches. The first involves pretraining models on existing datasets before generalizing to the specific domain of interest. The second strategy is to create synthetic datasets specifically tailored to the target domain using methods like copy-paste techniques or ad-hoc simulators. The first strategy often faces a significant domain shift, while the second demands custom solutions crafted for the specific task. In response to these challenges, here we propose a transfer learning framework that is valid for a generic scenario. In this framework, generated images help to improve the performances of an object detector in a few-real data regime. This is achieved through a diffusion-based generative model that was pretrained on large generic datasets. With respect to the state-of-the-art, we find that it is not necessary to fine tune the generative model on the specific domain of interest. We believe that this is an important advance because it mitigates the labor-intensive task of manual labeling the images in object detection tasks. We validate our approach focusing on fishes in an underwater environment, and on the more common domain of cars in an urban setting. Our method achieves detection performance comparable to models trained on thousands of images, using only a few hundreds of input data. Our results pave the way for new generative AI-based protocols for machine learning applications in various domains, for instance ranging from geophysics to biology and medicine. The provided datasets are built with the help of Gligen and the already existing NuImages, Ozfish and Deepfish datasets. The file "CarGenerated.zip" contains images generated with Gligen and with provided bounding boxes around cars in an urban environment. The file "fishes_on_bkg.zip" provides fish images generated with fishes from Deepfish inpainted with Gligen on generated backgrounds. The file "fish_text.zip" contains images completely generated with Gligen containing fishes with annotated bounding boxes. Finally, the file "oz_masked_512.zip" contains a simpler dataset of copy paste images of Deepfish fishes on Ozfish backrounds. All the files contains the images saved in different folders for training and validation, plus an index file called gt_fish.csv for the bounding boxes.

Figure 8. Prototype Front-End-Generative Learning Objects Instantiated with Random Numbers Based Expressions

We modeled classes for each AGLO section: scenario, theory, questions, and feedbacks. We modeled domain specific classes for the learned concepts like trees and graphs having generative methods controllable through parameters. Figure 8 depicts our prototype front-end.

Figure 7. Generated Tree Example-Generative Learning Objects Instantiated with Random Numbers Based Expressions

In this subsection we give an example of a model instantiation. We describe the instantiation process for all the symbols in the AGLO model. The nNoOfElements symbol was instantiated with the generated value of 10. The tabKeys symbol was instantiated with the [A,B,C,D,E,F,G,H,I] generated array. The tree symbol was instantiated with a ”Tree” object built with arguments equal with previously instantiated symbols. Class ”Tree” is part of the auxiliary JavaScript library.The treeSVG symbol was assigned to the tree SVG string representation obtained by a method call and is depicted in figure 7.

Figure 5. AGLO Scenario Symbols-Generative Learning Objects Instantiated with Random Numbers Based Expressions

analyzed AGLO that is displayed to the learner for localization and selection purposes. The second XML element is the scenario element containing a text description of the AGLO and a set of symbols. The description is expressed in natural language and we can notice that it contains four main steps: i) random tree generation; ii) index computation for presentation; iii) parent index computation for answer validation; iv) access to the first two keys for particular feedback generation. In the scenario section depicted in figure 5 several symbols are defined with the following semantics.

Figure 6. AGLO Presentation Sections-Generative Learning Objects Instantiated with Random Numbers Based Expressions

The seventh and eighth symbols named node1 and node2, both of type character, are initialized with the first two values from the keys array. These symbols will be used later in the feedback section. Their initialization is based on accessing the keys using the array indexing operator. Next, we discuss about the presentation sections of the AGLO depicted in figure 6.

Figure 3. Autogenerative Learning Object Model Definition (2)Generative Learning Objects Instantiated with Random Numbers Based Expressions

Formally, the theory section (see line 11) contains only static data. Here the educator places theoretical information about the learned concept in HTML format. The question, answers and feedbacks sections (see lines 12-13, 14-17, 18-21) contain a mix of static data and dynamic values referring the previously computed symbol values. The semantic of these sections is to create a dynamic content to be presented to the learner. Formalisms for specific details are also present, like: i) correctness - because we need to know which is the correct answer if they are multiple; ii) feedbacks answer identifier list - because feedbacks may be related to certain answers and showed to the learner only in certain conditions.

Figure 2. Autogenerative Learning Object Model Definition (1)-Generative Learning Objects Instantiated with Random Numbers Based Expressions

The proposed model is defined using the EBNF meta-language resulting in the concise grammar from figures 2 and 3. We will not present the full XML Schema because of clarity and space reasons. Structurally, the AGLO model has six sections: i) name; ii) scenario; iii) theory; iv) question; v) answers; vi) feedbacks. Next, we will explain each section in details.