We lifted 3 functions out of this paper's own repositories and ran 2 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 |
|---|---|---|
| VisionLearningGroup/Text-to-Clip_Retrieval | canonical | 2 of 3 |
| Function | Status | Where it lives |
|---|---|---|
| interpolated_prec_rec | Ran | VisionLearningGroup/Text-to-Clip_Retrieval/experiments/Text_to_Clip/test_fast/utils.py code served (permissive licence) · get_code("81aab88321de62c2") |
| segment_iou | Ran | VisionLearningGroup/Text-to-Clip_Retrieval/experiments/Text_to_Clip/test_fast/utils.py code served (permissive licence) · get_code("d0744a2fe3151508") |
| py_cpu_nms | Not yet run | VisionLearningGroup/Text-to-Clip_Retrieval/lib/nms/py_cpu_nms.py code served (permissive licence) · get_code("0087afa9a6463790") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
We address the problem of text-based activity retrieval in video. Given a sentence describing an activity, our task is to retrieve matching clips from an untrimmed video. To capture the inherent structures present in both text and video, we introduce a multilevel model that integrates vision and language features earlier and more tightly than prior work. First, we inject text features early on when generating clip proposals, to help eliminate unlikely clips and thus speed up processing and boost performance. Second, to learn a fine-grained similarity metric for retrieval, we use visual features to modulate the processing of query sentences at the word level in a recurrent neural network. A multi-task loss is also employed by adding query re-generation as an auxiliary task. Our approach significantly outperforms prior work on two challenging benchmarks: Charades-STA and ActivityNet Captions.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("1804.05113")
get_code_for_paper("1804.05113")
have("1804.05113")
Connect an agent — have() is free.