SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.01885 · CVPR · 2024

From Audio to Photoreal Embodiment: Synthesizing Humans in Conversations

Timur Bagautdinov, Shaojie Bai, Trevor Darrell, Angjoo Kanazawa, Alexander Richard, Javier Romero, Evonne Ng

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 18 functions out of this paper's own repositories and ran 15 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.

RepositoryRoleRan
facebookresearch/audio2photoreal canonical 15 of 18
FunctionStatusWhere it lives
default Ran facebookresearch/audio2photoreal/model/vqvae.py
pointer only (licence: NONE) · get_code("7c49b92babb27a7d")
approx_standard_normal_cdf Ran facebookresearch/audio2photoreal/diffusion/losses.py
pointer only (licence: NONE) · get_code("cfd76fd0d89574a4")
betas_for_alpha_bar Ran facebookresearch/audio2photoreal/diffusion/gaussian_diffusion.py
pointer only (licence: NONE) · get_code("2ab2316ac6fdd869")
collate_v2 Ran facebookresearch/audio2photoreal/data_loaders/tensors.py
pointer only (licence: NOASSERTION) · get_code("d84976cfde065645")
discretized_gaussian_log_likelihood Ran facebookresearch/audio2photoreal/diffusion/losses.py
pointer only (licence: NONE) · get_code("cd33283d615fb3d7")
get_model_args Ran facebookresearch/audio2photoreal/utils/model_util.py
pointer only (licence: NOASSERTION) · get_code("3596c18df070c4af")
get_named_beta_schedule Ran facebookresearch/audio2photoreal/diffusion/gaussian_diffusion.py
pointer only (licence: NONE) · get_code("97cb206ca0491baf")
get_param_groups_and_shapes Ran facebookresearch/audio2photoreal/diffusion/fp16_util.py
pointer only (licence: NONE) · get_code("e41367ad14ff58fd")
get_person_num Ran facebookresearch/audio2photoreal/utils/model_util.py
pointer only (licence: NOASSERTION) · get_code("315803aeb6b9f726")
laplace_smoothing Ran facebookresearch/audio2photoreal/model/vqvae.py
pointer only (licence: NONE) · get_code("c360b5ccff704a88")
make_beta_schedule Ran facebookresearch/audio2photoreal/model/utils.py
pointer only (licence: NONE) · get_code("c984209a2c66afbe")
make_master_params Ran facebookresearch/audio2photoreal/diffusion/fp16_util.py
pointer only (licence: NONE) · get_code("e20dd5102da3b050")
normal_kl Ran facebookresearch/audio2photoreal/diffusion/losses.py
pointer only (licence: NONE) · get_code("cf2798b666b231ca")
prob_mask_like Ran facebookresearch/audio2photoreal/model/utils.py
pointer only (licence: NONE) · get_code("3c6433dd421724e0")
unflatten_master_params Ran facebookresearch/audio2photoreal/diffusion/fp16_util.py
pointer only (licence: NONE) · get_code("64fff1e30802b815")
collate_tensors Not yet run facebookresearch/audio2photoreal/data_loaders/tensors.py
pointer only (licence: NONE) · get_code("0b9f005ecb737755")
extract Not yet run facebookresearch/audio2photoreal/model/utils.py
pointer only (licence: NONE) · get_code("09c8479d9a5b3e06")
lengths_to_mask Not yet run facebookresearch/audio2photoreal/data_loaders/tensors.py
pointer only (licence: NONE) · get_code("80c46a6e9bfa68f7")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Figure 1. Synthesizing photoreal conversational avatars. Given the audio from a dyadic conversation, we generate realistic conversational motion for the face, body, and hands. The motion can then be rendered as a photorealistic video. Please see results video.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2401.01885")
get_code_for_paper("2401.01885")
have("2401.01885")

Connect an agent — have() is free.