Research map / Detection and segmentation / Open-set and few-shot detection
Unsupervised Image Representation Learning with Deep Latent Particles
Tal Daniel, Aviv Tamar
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use DLP with 30 particles instead of KeyNet with 30 keypoints | Unsupervised landmark regression · MAFL | MSE (%): 2.58 -> 2.56 | Improved |
| Use DLP with 50 particles instead of KeyNet with 50 keypoints | Unsupervised landmark regression · MAFL | MSE (%): 2.54 -> 2.43 | Improved |
| Add log-variance to the regression features (ablation) | DLP with 25 particles · MAFL | MSE (%): 2.87 -> 2.52 | Improved |
| Add log-variance to the regression features (ablation) | DLP with 30 particles · MAFL | MSE (%): 2.56 -> 2.49 | Improved |
| Add learned particle features to the regression features (ablation) | DLP with 30 particles · MAFL | MSE (%): 2.56 -> 2.36 | Improved |
| Add learned particle features to the regression features (ablation) | DLP with 50 particles · MAFL | MSE (%): 2.43 -> 2.39 | Hurt |
| Use Chamfer-KL instead of Standard KL (ablation) | DLP with constant Gaussian prior · MAFL | MSE (%): 3.18 -> 3.07 | Improved |
| Use Chamfer-KL instead of Standard KL (ablation) | DLP with uniform prior proposals · MAFL | MSE (%): 3.39 -> 2.99 | Improved |
| Use Chamfer-KL instead of Standard KL (ablation) | DLP with SSM prior · MAFL | MSE (%): 3.24 -> 2.65 | Improved |
| Use the SSM prior instead of a constant Gaussian prior (ablation) | DLP with Chamfer-KL · MAFL | MSE (%): 3.07 -> 2.65 | Improved |
| Use the SSM prior instead of uniform prior proposals (ablation) | DLP with Chamfer-KL · MAFL | MSE (%): 2.99 -> 2.65 | Improved |
Closest papers by meaning
- 3D-DLP: Self-supervised 3D Object-centric Scene Representation Learning (ICML 2026 · similarity 0.89)
- Latent Convolutional Models (ICLR 2019 · similarity 0.85)
- Unsupervised Robust Disentangling of Latent Characteristics for Image Synthesis (ICCV 2019 · similarity 0.84)
- Differentiable and Stable Long-Range Tracking of Multiple Posterior Modes (NeurIPS 2023 · similarity 0.84)
- Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics Modeling (ICLR 2026 · similarity 0.84)
- Learning 3D Particle-based Simulators from RGB-D Videos (ICLR 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
