Xiang Bai, Wenjing Zhang, Xuhui Zhu
We lifted 15 functions out of this paper's own repositories and ran 0 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 |
|---|---|---|
| tiny-smart/OMAN | canonical | 0 of 15 |
| Function | Status | Where it lives |
|---|---|---|
| build_dataset | Not yet run | tiny-smart/OMAN/datasets/Sense_dataset.py code served (permissive licence) · get_code("0628e16af7cd6425") |
| build_decoder | Not yet run | tiny-smart/OMAN/models/transformer/prog_win_transformer.py code served (permissive licence) · get_code("98ae47001db9858f") |
| build_encoder | Not yet run | tiny-smart/OMAN/models/transformer/prog_win_transformer.py code served (permissive licence) · get_code("9922dcf47dc78e60") |
| build_matcher | Not yet run | tiny-smart/OMAN/models/matcher.py code served (permissive licence) · get_code("0b0a142826b76681") |
| build_video_dataset | Not yet run | tiny-smart/OMAN/datasets/Sense_dataset.py code served (permissive licence) · get_code("325f0ed27110ec3d") |
| compute_density | Not yet run | tiny-smart/OMAN/datasets/Locator_dataset.py code served (permissive licence) · get_code("f99cfa4658931193") |
| compute_relative_position | Not yet run | tiny-smart/OMAN/models/vic.py code served (permissive licence) · get_code("65ba998e1916b39a") |
| duplicate | Not yet run | tiny-smart/OMAN/models/pet.py code served (permissive licence) · get_code("ee3bda7721d1cc1f") |
| get_group_ones | Not yet run | tiny-smart/OMAN/models/tri_sim_ot_b.py code served (permissive licence) · get_code("a11ee826bb62d41a") |
| inverse_normalize | Not yet run | tiny-smart/OMAN/datasets/Sense_dataset.py code served (permissive licence) · get_code("14c1b21eca3aaeb9") |
| kl_loss | Not yet run | tiny-smart/OMAN/models/extra_loss.py code served (permissive licence) · get_code("2729a1dad2349aa9") |
| locator_dataset | Not yet run | tiny-smart/OMAN/datasets/Locator_dataset.py code served (permissive licence) · get_code("98f43ab84b89f23d") |
| pos2posemb2d | Not yet run | tiny-smart/OMAN/models/vic.py code served (permissive licence) · get_code("573b40489f6a85ca") |
| process_data | Not yet run | tiny-smart/OMAN/datasets/Locator_dataset.py code served (permissive licence) · get_code("3d064faa02210672") |
| similarity_cost | Not yet run | tiny-smart/OMAN/models/tri_sim_ot_b.py code served (permissive licence) · get_code("9f709b995cfb779e") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Video Individual Counting (VIC) is a recently introduced task aiming to estimate pedestrian flux from a video. It extends Video Crowd Counting (VCC) beyond the per-frame pedestrian count. In contrast to VCC that learns to count pedestrians across frames, VIC must identify coexistent pedestrians between frames, which turns out to be a correspondence problem. Existing VIC approaches, however, can underperform in congested scenes such as metro commuting. To address this, we build WuhanMetroCrowd, one of the first VIC datasets that characterize crowded, dynamic pedestrian flows. It features sparse-to-dense density levels, short-to-long video clips, slow-to-fast flow variations, front-to-back appearance changes, and light-to-heavy occlusions. To better adapt VIC approaches to crowds, we rethink the nature of VIC and recognize two informative priors: i) the social grouping prior that indicates pedestrians tend to gather in groups and ii) the spatial-temporal displacement prior that informs an individual cannot teleport physically. The former inspires us to relax the standard oneto-one (O2O) matching used by VIC to one-to-many (O2M) matching, implemented by an implicit context generator and a O2M matcher; the latter facilitates the design of a displacement prior injector, which strengthens not only O2M matching but also feature extraction and model training. These designs jointly form a novel and strong VIC baseline OMAN++. Extensive experiments show that OMAN++ not only outperforms stateof-the-art VIC baselines on the standard SenseCrowd, CroHD, and MovingDroneCrowd benchmarks, but also indicates a clear advantage in crowded scenes, with a 38.12% error reduction on our WuhanMetroCrowd dataset. Code, data, and pretrained models are available at https://github.com/tiny-smart/OMAN.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2601.01192")
get_code_for_paper("2601.01192")
have("2601.01192")
Connect an agent — have() is free.