TechcraftingAI Computer Vision

By Brad Edwards

TechcraftingAI Computer Vision brings you summaries of the latest arXiv research daily. Research is read by your virtual host, Sage. The podcast is produced by Brad Edwards, an AI Engineer from Vancou... more

· Technology

Download on the App Store

Download on the App Store

Get it on Google Play

FAQs about TechcraftingAI Computer Vision:

How many episodes does TechcraftingAI Computer Vision have?

The podcast currently has 315 episodes available.

TechcraftingAI Computer Vision episodes:

November 27, 2023Ep. 45 - November 23, 2023
Computer vision and pattern recognition research from arXiv for November 23, 2023.

Today's Themes (AI Generated)
Image manipulation and forgery detection methods and benchmarks

Efficient neural network compression techniques like pruning and reduced encodings

Text and language conditioning of generative models for controllable image synthesis

Robustifying vision systems, like CLIP, to image corruptions from compression

Exploiting multi-modality and cross-modality, especially vision and language, for representation learning
...more
57min
November 23, 2023Ep. 44 - November 22, 2023
Computer vision and pattern recognition research from arXiv for November 22, 2023.

Today's Themes (AI Generated)
3D reconstruction and novel view synthesis using neural radiance fields and differentiable rendering.

Human-AI collaboration for multi-rater learning with noisy labels.

Millimeter wave sensing for 3D object characterization and environment mapping.

Applications of spiking neural networks in visual place recognition.

Transformer-based models for tasks like image matting, copy-move forgery detection, semantic segmentation, and layout generation.
...more
1h
November 22, 2023Ep. 43 - November 21, 2023
Computer vision and pattern recognition research from arXiv for November 21, 2023.

Today's Themes (AI Generated)
Leveraging unlabeled data and self-supervision for semi-supervised learning and domain adaptation

Implicit neural 3D shape representations for reconstruction and modeling

Multi-modal perception with vision and language for navigation, instruction following, and question answering

Efficient architectures and techniques for deployment on mobile and embedded devices

Addressing model bias, robustness, and interpretability for trustworthy AI
...more
1h 7min
November 21, 2023Ep. 42 - November 20, 2023
Computer vision and pattern recognition research from arXiv for November 20, 2023.

Today's Themes (AI Generated)
Image restoration through diffusion models and transformer architectures

3D reconstruction and novel view synthesis with neural radiance fields

Facial analysis including emotion, age, and action unit recognition

Document understanding for invoices, recipes, and other complex layouts

Video prediction and generation using convolutional neural networks
...more
1h 29min
November 21, 2023Ep. 41 - November 19, 2023
Computer vision and pattern recognition research from arXiv for November 17, 2023.

Today's Themes (AI Generated)
Open vocabulary segmentation and detection using vision-language models

Generating imaginative images from text descriptions and storylines

Land cover mapping using sub-meter resolution imagery and human-guided learning

Enhancing robustness of vision-language models against adversarial image perturbations

Restoring faces occluded by masks for improved recognition
...more
38min
November 21, 2023Ep. 40 - November 18, 2023
Computer vision and pattern recognition research from arXiv for November 17, 2023.

Today's Themes (AI Generated)
Developing more robust visual representations, including through self-supervised learning, to improve model generalization capabilities

Applying diffusion models for enhanced image generation, including techniques to improve result diversity and plausibility

Leveraging transformer architectures for multi-modal and hierarchical analysis in medical imaging and biometrics

Studying adversarial attacks and defenses, evaluating real-time performance to enable vision applications in safety-critical domains

Analyzing model uncertainty for critical vision tasks like segmentation and reconstruction, to improve automation pipelines
...more
47min
November 20, 2023Ep. 39 - November 17, 2023
Computer vision and pattern recognition research from arXiv for November 17, 2023.

Today's Themes (AI Generated)
Image synthesis from text prompts and layouts

Object detection in domain adaptation settings

Video analysis using self-supervision and active learning

Medical image segmentation with limited annotations

Analysis of textures and celestial objects
...more
46min
November 17, 2023Ep. 38 - November 16, 2023
arXiv research summaries for Computation Vision and Pattern Recognition from November 16, 2023.

Today's Themes (AI Generated)
Improving domain adaptation techniques and image quality for diffusion models

Leveraging sparse annotations and weak supervision for segmentation tasks

Enhancing text-to-image generation with better consistency and fidelity

Developing efficient and calibrated neural radiance fields for 3D reconstruction

Applying transformers and self-supervision to improve downstream vision tasks
...more
53min
November 16, 2023Ep. 37 - November 15, 2023
arXiv research summaries for Computation Vision and Pattern Recognition from November 15, 2023.

Today's Themes (AI Generated)
Improving image and video quality with super resolution and denoising models.

Leveraging large language models for vision-language tasks like image captioning and visual question answering.

Developing unsupervised and self-supervised methods for specialized domains with limited labeled data.

Advancing autonomous vehicle perception through object detection, depth estimation and other computer vision techniques.

Applying diffusion models for image and 3D shape generation, reconstruction, and editing.
...more
55min
November 15, 2023Ep. 36 - November 14, 2023
arXiv research summaries for Computation Vision and Pattern Recognition from November 14, 2023.

Today's Themes (AI Generated)
Detection and segmentation of objects in challenging environments like sea clutter or histopathology images.

Applications of transformer architectures for tasks like video frame interpolation, action recognition, and graph matching.

Leveraging language models for multimodal tasks like image/video captioning and activity recognition from IoT sensor data.

Adversarial robustness of neural networks and connections to sparse coding.

Few-shot and zero/one-shot learning for generation and recognition tasks.
...more
47min

FAQs about TechcraftingAI Computer Vision:

How many episodes does TechcraftingAI Computer Vision have?

The podcast currently has 315 episodes available.