Online house building tools are robust instruments that have transformed the approach to home planning and design. The significance of 3D models on online house building and design platforms lies in their ability to elevate the user experience, enhance design precision, and foster collaboration. The creation of 3D house models is a demanding and detail-oriented undertaking that requires significant dedication and expertise. This thesis plans to investigate possibilities of implementing AI with a limited amount of data, to help streamline the manual process of working on 3D models for use in online house building tools. Multiple problems arouse related to issues with the acquired house mesh data, issues with the chosen AI models and hardware issues.  The problems during the different stages of work made any meaningful analyses hard but some evaluations could still be extracted. Time was a central aspect in that the time taken to train the AI models greatly correlated with the amount of data used for said training. It was also made apparent that there was a great need for more data if the AI was to be able to be trained on a dataset made from 3D house meshes. If this type of project was to be picked up in the future, a recommendation would be to not go at it alone as the time needed to perform the different steps should not be underestimated.

Mitt projekt är en animerad kortfilm där vi får följa ekollonet Acca och hennes dagliga rutiner. Vi ser henne gå från ung till gammal och hur förutsättningarna runt henne förändras. Examensarbetet är också ett utforskande för mig i en nyfunnen teknik och även ett undersökandeav hur en visuell kreatör kan hålla sin process mer hållbar och klimatsmart. / An animated short film in which we follow the acorn Acca through her daily routines. We watch her go from young to old and observe how the conditions around her change. The degree project is also an exploration for me in a newfound technique, as well as an examination of how a visual creator can keep their process more sustainable and climate-smart.

Homogeneity of powder blends is an important metric for industrial applications in fortified dairy foods including commercial pediatric nutrition products. Product development practices evaluating physical properties and nutrient stability performance are reliant on blending parameters that deliver a uniform powder at both the pilot and commercial scale. Quantities of individual micronutrients in finished products are particularly critical for formulated infant foods. The two preliminary phases of this study focused on developing a simple, efficient method, specifically for a pilot scale ribbon blender, in which maximum homogeneity in fortified dairy-based powder blends could be reached. In phase one, a red iron oxide pigment powder was mixed throughout a white dairy powder and color homogeneity was measured by comparing L*a*b* color values from powder samples extracted from different areas of the ribbon blender. For phase two, sixteen similar fortified dairy blends were produced with varying ribbon blade shaft rpm, fill level, and blending durations according to a response surface method (RSM). The level of homogeneity of ferrous sulfate in the dairy blend was measured in these fortified mixtures to determine optimal blender parameters. After operating parameters were determined for uniform blending, phase three was enacted using these parameters. A comparison study of nutrient stability in fortified model non-agglomerated powder infant formula (PIF) and agglomerated whey protein concentrate (WPC) powder blends was executed to evaluate the degradative effect of microencapsulated ferrous sulfate (MFS) vs. unencapsulated ferrous sulfate (UFS) in these fortified dairy blends. The nutrient degradation rates of vitamins A, E and C in both PIF and WPC base powder, fortified with either MFS or UFS, were determined and compared during an accelerated eight-week stability study. Using p = 0.05, no statistically significant differences in vitamin degradation rates were observed when comparing independent spray-dried dairy-based blends containing unencapsulated or microencapsulated ferrous sulfate (using an encapsulating composition of 60% stearic acid) during eight weeks of accelerated shelf-life storage conditions (37 °C with a 75% relative humidity, RH). Of note, the degradation rates of vitamins A and E in blends containing PIF and UFS were more rapid than the control and suggestively significantly different (p = 0.07).

No description available.

Syftet med denna undersökning är att bidra med djupare förståelse för hur rörligt material uppfattas beroende på dels var betraktaren befinner sig och dels om visuellt material förvrängs eller ej. Detta genomfördes genom ett test och efterföljande intervjuer med åtta individer som studerade medieproduktion vid Högskolan Dalarna. Deltagarna tillfrågades vilka skillnader de upplevde mellan de olika vinklarna, vilken som kändes mest realistisk, om de upplevde att materialet ändrades beroende på vilken vinkel de såg och om de upplevde att de olika vinklarna var verklighetstrogna. Resultatet visade en delad uppfattning huruvida 2D videos förvrängdes eller ej när de beskådades från olika betraktningsplatser. Beroende på betraktningsplatsen så uppfattades 2D videos som både normala och förvrängda. Sammanfattningsvis kan inga direkta slutsatser om huruvida 2D videos uppfattas när det beskådas från olika betraktningsplatser eller vilken roll betraktningsplatsen spelar för uppfattningen av 2D videos dras. Svagheten i undersökningen är att den endast intervjuat personer som studerar eller arbetar med medieproduktion och att det komplexa materialet upplevdes simpelt. Förslag till fortsatt forskning är således att undersöka mer komplexa animationer och intervjua ett större antal personer med olika kunskaper i förhållande till medieproduktion.

Facade cracks are a common problem in the north of Sweden due to shifting temperatures creating frost in the facades which ultimately damages the facades, often in the form of cracks. To fix these cracks, workers must visually inspect the facades to find them which is a difficult and time-consuming task. This project explores the possibilities of creating an algorithm that can classify cracks on facades with the help of deep learning models. The idea is that in the future, an algorithm like this could be implemented on a drone that hoovers around buildings, filming the facade, and reporting back if there are any damages to the facade. The work in this project is exploratory and the path of convolutional neural networks has been explored, as well as the possibility to simulate training data due to the lack of real-world data. The experimental work in this project led to some interesting conclusions for further work. The relatively small amount of data used in this project points towards the possibility of using simulated data as a complement to real data, as well as the possibility of using convolutional neural networks as a means of classifying facades for crack recognition. The data and conclusions collected in this report can be used as a preparatory work for a working prototype algorithm.

Common Machine-Learning (ML) approaches for scene classification require a large amountof training data. However, for classification of depth sensor data, in contrast to image data, relativelyfew databases are publicly available and manual generation of semantically labeled 3D point clouds isan even more time-consuming task. To simplify the training data generation process for a wide rangeof domains, we have developed theBLAINDERadd-on package for the open-source 3D modelingsoftware Blender, which enables a largely automated generation of semantically annotated point-cloud data in virtual 3D environments. In this paper, we focus on classical depth-sensing techniquesLight Detection and Ranging (LiDAR) and Sound Navigation and Ranging (Sonar). Within theBLAINDERadd-on, different depth sensors can be loaded from presets, customized sensors can beimplemented and different environmental conditions (e.g., influence of rain, dust) can be simulated.The semantically labeled data can be exported to various 2D and 3D formats and are thus optimizedfor different ML applications and visualizations. In addition, semantically labeled images can beexported using the rendering functionalities of Blender.

<p>This thesis presents an approach for automatically generating signing animations from a sign language notation. An avatar endowed with expressive gestures, as subtle as changes in facial expression, is used to render the sign language animations. SWML, an XML format of SignWriting is provided as input. It transcribes sign language gestures in a format compatible to virtual signing. Relevant features of sign language gestures are extracted from the SWML. These features are then converted to body animation pa- rameters, which are used to animate the avatar. Using key-frame animation techniques, intermediate key-frames approximate the expected sign language gestures. The avatar then renders the corresponding sign language gestures. These gestures are realistic and aesthetically acceptable and can be recognized and understood by Deaf people.</p>

<p>This thesis presents an approach for automatically generating signing animations from a sign language notation. An avatar endowed with expressive gestures, as subtle as changes in facial expression, is used to render the sign language animations. SWML, an XML format of SignWriting is provided as input. It transcribes sign language gestures in a format compatible to virtual signing. Relevant features of sign language gestures are extracted from the SWML. These features are then converted to body animation pa- rameters, which are used to animate the avatar. Using key-frame animation techniques, intermediate key-frames approximate the expected sign language gestures. The avatar then renders the corresponding sign language gestures. These gestures are realistic and aesthetically acceptable and can be recognized and understood by Deaf people.</p>

For years, game companies have been using different programs for the creation of the newest most inspiring games. A short presentation about different programs used for game art is provided with key aspects in developing art for games in character production. One aim is to showcase the potential of open-source programs. Two projects with two different characters were created for games testing the workflow between open-source, free application and also commercial programs. The results of the projects are discussed and then the document focuses on explaining what the negative side of open-source programs is and why they should not be used. After the discussion, a conclusion is made explaining the similarities and differences between open-source and commercial programs.