We lifted 10 functions out of this paper's own repositories and ran 9 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| RonMcKay/OODRetrieval | canonical | 9 of 10 |
| Function | Status | Where it lives |
|---|---|---|
| cos_dist | Ran | RonMcKay/OODRetrieval/eval_retrieval.py code served (permissive licence) · get_code("f7c25fd44cb0994d") |
| feature_resnet101 | Ran | RonMcKay/OODRetrieval/src/embedding_networks.py code served (permissive licence) · get_code("25d3a44152ac73b9") |
| feature_resnet18 | Ran | RonMcKay/OODRetrieval/src/embedding_networks.py code served (permissive licence) · get_code("d33581d9a4dc7c99") |
| feature_vgg16 | Ran | RonMcKay/OODRetrieval/src/embedding_networks.py code served (permissive licence) · get_code("48e466ac44f9532d") |
| get_gt | Ran | RonMcKay/OODRetrieval/eval_selection.py code served (permissive licence) · get_code("764c98574dd295b3") |
| get_image_index_to_components | Ran | RonMcKay/OODRetrieval/compute_embeddings.py code served (permissive licence) · get_code("a3ad7fe4cf010b25") |
| iou_numpy | Ran | RonMcKay/OODRetrieval/src/eval_utils.py code served (permissive licence) · get_code("a3998eac9d647345") |
| lp_dist | Ran | RonMcKay/OODRetrieval/eval_retrieval.py code served (permissive licence) · get_code("4d2f371cb844db06") |
| return_and_update_instances | Ran | RonMcKay/OODRetrieval/eval_selection.py code served (permissive licence) · get_code("2af2fd60326f6f8a") |
| ap_wrapper | Not yet run | RonMcKay/OODRetrieval/eval_retrieval.py code served (permissive licence) · get_code("4ddf7b8214292152") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
When deploying deep learning technology in self-driving cars, deep neural networks are constantly exposed to domain shifts. These include, e.g., changes in weather conditions, time of day, and long-term temporal shift. In this work we utilize a deep neural network trained on the Cityscapes dataset containing urban street scenes and infer images from a different dataset, the A2D2 dataset, containing also countryside and highway images. We present a novel pipeline for semantic segmenation that detects out-of-distribution (OOD) segments by means of the deep neural network's prediction and performs image retrieval after feature extraction and dimensionality reduction on image patches. In our experiments we demonstrate that the deployed OOD approach is suitable for detecting out-of-distribution concepts. Furthermore, we evaluate the image patch retrieval qualitatively as well as quantitatively by means of the semi-compatible A2D2 ground truth and obtain mAP values of up to 52.2%.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2005.06831")
get_code_for_paper("2005.06831")
have("2005.06831")
Connect an agent — have() is free.