Papers › CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters
CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters
Paul Gavrikov, Janis Keuper
Currently, many theoretical as well as practically relevant questions towards the transferability and robustness of Convolutional Neural Networks (CNNs) remain unsolved. While ongoing research efforts are engaging these problems from various angles, in most computer vision related cases these approaches can be generalized to investigations of the effects of distribution shifts in image data. In this context, we propose to study the shifts in the learned weights of trained CNN models. Here we focus on the properties of the distributions of dominantly used 3x3 convolution filter kernels. We collected and publicly provide a dataset with over 1.4 billion filters from hundreds of trained CNNs, using a wide range of datasets, architectures, and vision tasks. In a first use case of the proposed dataset, we can show highly relevant properties of many publicly available pre-trained models for practical applications: I) We analyze distribution shifts (or the lack thereof) between trained filters along different axes of meta-parameters, like visual category of the dataset, task, architecture, or layer depth. Based on these results, we conclude that model pre-training can succeed on arbitrary datasets if they meet size and variance conditions. II) We show that many pre-trained models contain degenerated filters which make them less robust and less suitable for fine-tuning on target applications. Data & Project website: https://github.com/paulgavrikov/cnn-filter-db
In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.
Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Datasets
Introduced by this paper, per the archive.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Image Classification | CIFAR-10 | ResNet-9 | Percentage correct | 94.79 | #148 of 265 | Archive leaderboard | report |
| Image Classification | CIFAR-100 | ResNet-9 | Percentage correct | 75.59 | #152 of 211 | Archive leaderboard | report |
| Image Classification | Fashion-MNIST | Inception v3 | Accuracy | 94.44 | #5 of 34 | Archive leaderboard | report |
| Image Classification | Fashion-MNIST | Inception v3 | Percentage error | 5.56 | #5 of 34 | Archive leaderboard | report |
| Image Classification | Kuzushiji-MNIST | ResNet-14 | Accuracy | 98.75 | #11 of 26 | Archive leaderboard | report |
| Image Classification | MNIST | ResNet-9 | Accuracy | 99.68 | #66 of 81 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections